The Ne w E n g l a nd J o u r n a l o f Me d ic i neMechanisms of DiseaseConditions that commonly lead to substantialchanges in the plasma concentrations of acute-phaseproteins include infection, trauma, surgery, burns,F R A N K L I N H . E P S T E I N , M . D. , Editortissue infarction, various immunologically mediatedand crystal-induced inflammatory conditions, andadvanced cancer. Moderate changes occur after stren-ACUTE-PHASE PROTEINS AND OTHER uous exercise, heatstroke, and childbirth. Small chang-SYSTEMIC RESPONSES TO es occur after psychological stress and in several psy-INFLAMMATIONchiatric illnesses.6 Although the concentrations ofmultiple components of the acute-phase responseCcommonly increase together, not all of them in-EM GABAY, M.D., AND IRVING KUSHNER, M.D.crease uniformly in all patients with the same illness.Thus, febrile patients may have normal plasma con- LARGE number of changes, distant fromcentrations of C-reactive protein, and discordancethe site or sites of inflammation and involv-Abetween the plasma concentrations of different acute-ing many organ systems, may accompany in-phase proteins is common. These variations, whichflammation. In 1930 interest was focused on theseindicate that the components of the acute-phase re-changes by the discovery of C-reactive protein (sosponse are individually regulated, may be explainednamed because it reacted with the pneumococcalin part by differences in the patterns of productionC-polysaccharide) in the plasma of patients duringof specific cytokines or their modulators in differentthe acute phase of pneumococcal pneumonia.1 Ac-pathophysiologic states.cordingly, these systemic changes have since been re-ferred to as the acute-phase response,2 even thoughREGULATION OF ACUTE-PHASE CHANGESthey accompany both acute and chronic inflamma-Induction of Acute-Phase Proteins by Cytokines tory disorders. New acute-phase phenomena contin-and Other Extracellular Signaling Moleculesue to be recognized, and the mechanisms mediatingCytokines are intercellular signaling polypeptidesthem are becoming better understood. This reviewproduced by activated cells. Most cytokines havesummarizes much of the knowledge that has beenmultiple sources, multiple targets, and multiple func-acquired about the acute-phase response since thistions. The cytokines that are produced during andsubject was last reviewed in the Journal in 1984.3participate in inflammatory processes are the chiefACUTE-PHASE RESPONSESstimulators of the production of acute-phase pro-teins. These inflammation-associated cytokines in-Acute-phase changes may be divided into changesclude interleukin-6, interleukin-1b, tumor necrosisin the concentrations of many plasma proteins, knownfactor a, interferon-g, transforming growth factoras the acute-phase proteins (Table 1), and a largeb,2 and possibly interleukin-8.7 They are producednumber of behavioral, physiologic, biochemical, andby a variety of cell types, but the most importantnutritional changes (Table 2). An acute-phase pro-sources are macrophages and monocytes at inflam-tein has been defined as one whose plasma concen-matory sites.tration increases (positive acute-phase proteins) or de-Interleukin-6 is the chief stimulator of the pro-creases (negative acute-phase proteins) by at least 25duction of most acute-phase proteins,8 whereas thepercent during inflammatory disorders.4 The chang-other implicated cytokines influence subgroups ofes in the concentrations of acute-phase proteins areacute-phase proteins. However, in mice rendered in-due largely to changes in their production by hepa-capable of expressing interleukin-6 (knockout mice),tocytes. The magnitude of the increases varies fromthe role of interleukin-6 in stimulating the produc-about 50 percent in the case of ceruloplasmin andtion of acute-phase proteins depends on the natureseveral complement components to as much asor site of the inflammatory stimulus; the response is1000-fold in the case of C-reactive protein and se-largely inhibited in interleukin-6 knockout mice in-rum amyloid A, the plasma precursor of amyloid Ajected with turpentine but is normal when bacterial(the principal constituent of secondary amyloid de-lipopolysaccharide is the inflammatory stimulus.9 Thisposits) (Fig. 1). finding indicates that lipopolysaccharide causes theproduction of other cytokines capable of stimulatingthe production of acute-phase proteins.9 The re-From the Division of Rheumatology, Department of Medicine, Univer-sponses are similar in interleukin-1b knockout mice,sity of Colorado Health Sciences Center, Denver (C.G.); and the Depart-presumably because interleukin-1b is required toments of Medicine and Pathology and the Rammelkamp Center for Re-search, Case Western Reserve University, MetroHealth Campus, Clevelandstimulate the production of interleukin-6 after the(I.K.). Address reprint requests to Dr. Kushner at the Department of Med-administration of turpentine.10 These studies provideicine, Case Western Reserve University at MetroHealth Medical Center,further evidence that patterns of cytokine produc-Cleveland, OH 44109-1998, or at [email protected]©1999, Massachusetts Medical Society.tion and the acute-phase response differ in different448·Fe b r u a r y 11 , 19 9 9Downloaded from www.nejm.org by VICTOR ARREDONDO MD on March 27, 2007 . Copyright © 1999 Massachusetts Medical Society. All rights reserved. M EC H A N I S M S O F D I S E AS ETABLE 1. HUMAN ACUTE-PHASE PROTEINS.TABLE 2. OTHER ACUTE-PHASE PHENOMENA.Proteins whose plasma concentrationsNeuroendocrine changesincreaseFever, somnolence, and anorexiaComplement systemIncreased secretion of corticotropin-releasing hormone, C3corticotropin, and cortisolC4Increased secretion of arginine vasopressinC9Decreased production of insulin-like growth factor IFactor BIncreased adrenal secretion of catecholaminesC1 inhibitorHematopoietic changesC4b-binding proteinAnemia of chronic diseaseMannose-binding lectinLeukocytosisCoagulation and fibrinolytic systemThrombocytosisFibrinogenMetabolic changesPlasminogenLoss of muscle and negative nitrogen balanceTissue plasminogen activatorDecreased gluconeogenesisUrokinaseOsteoporosisProtein SIncreased hepatic lipogenesisVitronectinIncreased lipolysis in adipose tissuePlasminogen-activator inhibitor 1Decreased lipoprotein lipase activity in muscle and adiposeAntiproteasestissuea -Protease inhibitorCachexia1a -AntichymotrypsinHepatic changes1Pancreatic secretory trypsin inhibitorIncreased metallothionein, inducible nitric oxide synthase, Inter-a-trypsin inhibitorsheme oxygenase, manganese superoxide dismutase, and Transport proteinstissue inhibitor of metalloproteinase-1CeruloplasminDecreased phosphoenolpyruvate carboxykinase activityHaptoglobinChanges in nonprotein plasma constituentsHemopexinHypozincemia, hypoferremia, and hypercupremiaParticipants in inflammatory responsesIncreased plasma retinol and glutathione concentrationsSecreted phospholipase A2Lipopolysaccharide-binding proteinInterleukin-1–receptor antagonistGranulocyte colony-stimulating factorOthersC-reactive proteinIn addition, cytokines are components of a large,Serum amyloid Aa -Acid glycoproteincomplex signaling network. Most likely, cells are sel-1Fibronectindom exposed to only a single cytokine. Instead,Ferritincombinations of mediators convey biologically rele-Angiotensinogenvant information. The effects of cytokines on targetProteins whose plasma concentrations decreasecells may be inhibited or enhanced by other cyto-Albuminkines, by hormones, and by cytokine-receptor an-Transferrintagonists and circulating receptors. Combinations ofTransthyretincytokines have been found to have additive, inhibi-a -HS glycoprotein2Alpha-fetoproteintory, or synergistic effects.15 Thus, the induction ofThyroxine-binding globulinC-reactive protein and serum amyloid A in someInsulin-like growth factor IFactor XIImodels requires both interleukin-6 and either inter-leukin-1 or tumor necrosis factor a, and the induc-tion of fibrinogen by interleukin-6 is inhibited byinterleukin-1, tumor necrosis factor a, and trans-forming growth factor b.15 Interleukin-6 enhancesinflammatory conditions. Interleukin-11, leukemia in-the effect of interleukin-1b in inducing the expres-hibitory factor, oncostatin M, ciliary neurotrophicsion of interleukin-1–receptor antagonist,16 and inter-factor, and cardiotrophin 1 may have actions similarleukin-4 inhibits the induction of some acute-phaseto those of interleukin-6.11proteins by other cytokines.17 Soluble interleukin-6Cytokines operate both as a cascade and as a net-receptor a molecules increase the effects of the lig-work in stimulating the production of acute-phaseand,18 whereas other soluble receptors, such as thoseproteins. Many cytokines can regulate the produc-for tumor necrosis factor a and interleukin-1, are in-tion of other cytokines and cytokine receptors. Forhibitory. Glucocorticoids generally enhance the stim-example, tumor necrosis factor a is the main stimu-ulatory effects of cytokines on the production oflator of interleukin-1 production in patients withacute-phase proteins,19 whereas insulin decreases theirrheumatoid arthritis12; interleukin-1b may increaseeffects on the production of some acute-phase pro-or decrease expression of its own receptors13; the in-teins.20terleukin-6 response to the injection of turpentineThe expression of genes for acute-phase proteinsin mice requires interleukin-1b; and interleukin-6is regulated mainly at the transcriptional level, butinhibits the expression of tumor necrosis factor a.14post-transcriptional mechanisms also participate.21,22Vo l u m e 3 4 0Nu m b e r 6·449Downloaded from www.nejm.org by VICTOR ARREDONDO MD on March 27, 2007 . Copyright © 1999 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd J o u r n a l o f Me d ic i ne30,100production, and also directly stimulate the adrenalgland. Stimulation of the production of arginine30,000vasopressin by interleukin-6 can explain the hypo-natremia that occurs during some inflammatory dis-700orders.C-reactive proteinThe behavioral changes that often accompany in-600flammation, including anorexia, somnolence, and500lethargy, are similarly induced by cytokines. NeuralSerum amyloid Amechanisms have also been implicated in anorexia;400as with fever, vagal afferents are required for the in-duction of anorexia after intraperitoneal injection of300interleukin-1b and lipopolysaccharide. Increased plas-HaptoglobinFibrinogenma leptin concentrations occur in inflammation,200probably in response to stimulation of adipocytes by100cytokines, and may also contribute to anorexia.30C3Inflammation-associated cytokines have been im-Change in Plasma Concentration (%)0plicated in the pathogenesis of anemia in chronicTransferrindisease; examples of their involvement include theAlbumindecreased responsiveness of erythrocyte precursors071421to erythropoietin, decreased production of erythro-Time after Inflammatory Stimulus (days)poietin, and impaired mobilization of iron from mac-Figure 1. Characteristic Patterns of Change in Plasma Concen-rophages.31 Hypoferremia results largely from thetrations of Some Acute-Phase Proteins after a Moderate Inflam-sequestration of iron in macrophages by apoferritinmatory Stimulus.produced in response to the inflammation-associat-Modified from Gitlin and Colten5 with the permission of theed cytokines interleukin-4 and interleukin-13.32 Thepublisher.thrombocytosis of inflammation appears to be causedby interleukin-6.33 Finally, cachexia, the loss of bodymass that occurs in severe chronic inflammatory dis-Post-translational changes in the glycosylation of plas-ease, results from decreases in skeletal muscle, fat tis-ma proteins during inflammatory states include al-sue, and bone mass. Interleukin-1b, interleukin-6,terations in oligosaccharide branching,23 increasedtumor necrosis factor a, and interferon g all contrib-sialylation of orosomucoid,24 and decreased galactos-ute to these processes.34-36ylation of IgG.25 Changes in oligosaccharide branch-Inflammation-associated cytokines also alter manying are induced by inflammation-associated cytokines,intracellular hepatic constituents, including induc-independently of their effects on the production ofible nitric oxide synthase, manganese superoxide dis-acute-phase proteins. Finally, the efficiency of secre-mutase, and microsomal heme oxygenase. Interleu-tion of C-reactive protein, a process distinct from itskin-6 increases the production of the metal-bindingsynthesis, is greatly increased during the acute-phaseprotein metallothionein, with consequent increasedresponse.26zinc binding and hypozincemia. Interleukin-1b andtumor necrosis factor a decrease the expression ofRegulation of Other Acute-Phase Changes growth-hormone receptors on hepatocytes, with sub-by Inflammation-Associated Cytokinessequent decreased responsiveness to growth hormoneFever is representative of the neuroendocrineand low plasma concentrations of insulin-like growthchanges that characterize the acute-phase response.factor I.37 Transgenic mice that overexpress interleu-Although several cytokines may induce fever, inter-kin-6 have low plasma concentrations of insulin-likeleukin-6 produced in the brain stem is required forgrowth factor I and are smaller than normal mice.the final steps leading to fever.27 However, cytokinesThis finding suggests that increased production ofare not the sole inducers of fever; the recent findinginflammation-associated cytokines may explain, atthat subdiaphragmatic vagotomy blocks fever afterleast in part, impaired growth in children with chron-intraperitoneal (but not intramuscular) injection ofic inflammatory conditions.38lipopolysaccharide implicates neural transmission inthe febrile response.28 Other neuroendocrine chang-POSTULATED FUNCTION OF THE es reflect complex interactions among cytokines, theACUTE-PHASE RESPONSEhypothalamic–pituitary–adrenal axis, and other com-The assumption that the changes in plasma con-ponents of the neuroendocrine system.29 For exam-centrations of acute-phase proteins are beneficial isple, inflammation-associated cytokines stimulate thebased largely on the known functional capabilities ofproduction of corticotropin-releasing hormone, withthe proteins and on logical speculation as to howconsequent stimulation of corticotropin and cortisolthese might serve useful purposes in inflammation,450·Fe b r u a r y 11 , 19 9 9Downloaded from www.nejm.org by VICTOR ARREDONDO MD on March 27, 2007 . Copyright © 1999 Massachusetts Medical Society. All rights reserved. M EC H A N I S M S O F D I S E AS Ehealing, or adaptation to a noxious stimulus. Inflam-In contrast, other acute-phase proteins may havemation is a complex, highly orchestrated process in-antiinflammatory actions. For example, the antioxi-volving many cell types and molecules, some ofdants haptoglobin and hemopexin protect againstwhich initiate, amplify, or sustain the process, somereactive oxygen species. Both a -protease inhibitor1of which attenuate it, and some of which cause it toand a -antichymotrypsin antagonize the activity of1resolve. A number of the participating molecules areproteolytic enzymes; a -antichymotrypsin also in-1multifunctional and contribute to both the waxinghibits the generation of superoxide anion.49 Woundand the waning of inflammation at different pointshealing can be influenced by two acute-phase pro-in its evolution.39teins: fibrinogen can cause endothelial-cell adhesion,Many of the acute-phase proteins have the poten-spreading, and proliferation, all critical to tissue re-tial to influence one or more of these stages of in-pair, and haptoglobin aids in wound repair by stim-flammation. A major function of C-reactive protein,ulating angiogenesis.50a component of the innate immune system, is itsIt is not clear what functional advantages mayability to bind phosphocholine and thus recognizearise from decreases in plasma concentrations of thesome foreign pathogens as well as phospholipid con-negative acute-phase proteins. It is logical to pre-stituents of damaged cells.40 It can activate the com-sume that the need to divert available amino acids toplement system when bound to one of its ligandsthe production of other acute-phase proteins ex-and can also bind to phagocytic cells, an observationplains the decreased production of plasma proteinssuggesting that it can initiate the elimination of tar-not required for host defense. Since transthyretin, ageted cells by its interaction with both humoral andnegative acute-phase protein, inhibits interleukin-1cellular effector systems of inflammation.40 Other pro-production by monocytes and endothelial cells,51 ainflammatory effects of C-reactive protein includedecrease in its plasma concentration may be pro-the induction of inflammatory cytokines and tissueinflammatory.factor in monocytes.41,42 However, its net effect isIt is possible to speculate about the beneficial ef-antiinflammatory in transgenic mice that producefects of other acute-phase phenomena. Somnolencelarge amounts of C-reactive protein.43,44 Such an ef-associated with inflammatory states may reduce de-fect of C-reactive protein may be explained by itsmands for energy. The adaptive value of fever hasability to prevent the adhesion of neutrophils to en-been attributed to both the enhancement of immu-dothelial cells by decreasing the surface expression ofnity and the stabilization of cell membranes.52 Glu-L-selectin,45 to inhibit the generation of superoxidecocorticoids have a major role in the maintenanceby neutrophils, and to stimulate the synthesis of in-of hemodynamic stability during severe illness andterleukin-1–receptor antagonist by mononuclear cells.can modulate the immune and inflammatory re-It seems likely that C-reactive protein has many patho-sponses to different noxious stimuli. The increase inphysiologic roles in the inflammatory process.plasma lipid concentrations may provide nutrientsThe actions of serum amyloid A, the other majorto cells involved in host defense and substrates foracute-phase protein in humans, are largely unknown.the regeneration of damaged membranes. In addi-Serum amyloid A consists of a family of apolipo-tion, circulating lipoproteins can bind to and de-proteins that rapidly bind to high-density lipopro-crease the toxic effects of lipopolysaccharide, thus par-tein after their synthesis and have the potential to in-ticipating in the defense against microbes.36 Increasedfluence cholesterol metabolism during inflammatoryproduction of heme oxygenase and manganese super-states.46,47 Serum amyloid A has also been reportedoxide dismutase may limit oxidant-mediated tissueto cause adhesion and chemotaxis of phagocytic cellsinjury, whereas tissue inhibitor of metalloprotein-and lymphocytes and may contribute to the inflam-ase-1 antagonizes the destructive effects of metallo-mation in atherosclerotic coronary arteries by in-proteinases.creasing the oxidation of low-density lipoproteins.48As with all inflammation-associated phenomena,Several acute-phase proteins initiate or sustain in-the acute-phase response is not uniformly beneficial.flammation. The classic complement components,Anemia and impaired growth have been mentionedmany of which are acute-phase proteins, have centralabove. When extreme, cytokine-induced changes as-proinflammatory roles in immunity, as does mannose-sociated with the acute-phase response can be fatal,binding lectin, a recently recognized acute-phaseas in septic shock.53 In addition, the persistence ofcomponent of complement. Complement activationthe acute-phase response because of long-term stim-leads to chemotaxis, plasma protein exudation at in-ulation, as in advanced cancer and the acquired im-flammatory sites, and opsonization of infectiousmunodeficiency syndrome, can lead to metabolicagents and damaged cells. Similarly, granulocyte col-disturbances that ultimately result in cachexia. Final-ony-stimulating factor increases the inflammatoryly, secondary amyloidosis has long been recognizedresponse by increasing the numbers of granulocyteas a deleterious consequence of elevated serum amy-precursors in bone marrow and by activating matureloid A concentrations in some patients with chronicgranulocytes.inflammatory conditions.Vo l u m e 3 4 0Nu m b e r 6·451Downloaded from www.nejm.org by VICTOR ARREDONDO MD on March 27, 2007 . Copyright © 1999 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd J o u r n a l o f Me d ic i neCLINICAL ASSESSMENT OF ACUTE-PHASE rocyte sedimentation rate increases steadily with age,PROTEINS AND CYTOKINESbut plasma C-reactive protein concentrations do not.Both anemia and hypoalbuminemia due to inflam-Systemic lupus erythematosus is one exception tomation are common among hospitalized patients.the generalization that C-reactive protein concentra-Estimation of other changes in acute-phase proteins,tions correlate with the extent and severity of inflam-despite the lack of diagnostic specificity, is useful tomation. Many patients with active systemic lupusclinicians because such changes reflect the presenceerythematosus do not have high plasma concentra-and intensity of an inflammatory process. Thus,tions of C-reactive protein (or serum amyloid A) butmeasurements of plasma or serum C-reactive pro-do have marked increases during bacterial infec-tein can help differentiate inflammatory from non-tion.55 Application of this knowledge to the differ-inflammatory conditions and are useful in managingential diagnosis of fever in patients with systemic lu-the patient’s disease, since the concentration oftenpus erythematosus has been somewhat limited byreflects the response to and need for therapeutic in-the finding that plasma C-reactive protein concen-tervention. Finally, in some diseases, such as rheu-trations are also high in patients with active lupusmatoid arthritis, serial measurements of C-reactiveserositis56 or chronic synovitis.57protein are of prognostic value.54 In evaluating lab-Most normal subjects have plasma C-reactive pro-oratory results, physicians should be aware thattein concentrations of 2 mg per liter or less, but somesome laboratories report C-reactive protein concen-have concentrations as high as 10 mg per liter. Thetrations in milligrams per liter and others in milli-latter finding, attributed to limited stimulation bygrams per deciliter. Serum amyloid A concentrationsminimally apparent low-grade processes such as gin-usually parallel those of C-reactive protein, althoughgivitis or trivial injury, has led to the suggestion thatsome studies indicate that serum amyloid A is avalues of less than 10 mg per liter should be regardedmore sensitive marker of inflammatory disease.46 Atas clinically unimportant.4 However, C-reactive pro-present, assays for serum amyloid A are not widelytein concentrations well below 10 mg per liter, butavailable.higher than those in most normal subjects, have beenCurrently, the most widely used indicators of thefound in patients with osteoarthritis, particularlyresponse of acute-phase proteins are the erythrocytethose with progressive joint damage.58 This observa-sedimentation rate and the plasma C-reactive pro-tion supports other data indicating the participationtein concentration. The rate at which erythrocytesof inflammation in this disorder. In addition, slightlyfall through plasma — that is, the erythrocyte sedi-elevated concentrations of C-reactive protein, withinmentation rate — depends largely on the plasmathe range in normal subjects, have been found to pre-concentration of fibrinogen. As a test, the erythro-dict subsequent coronary events, often years later, incyte sedimentation rate has the advantages of famil-patients with angina and in healthy physicians.59,60iarity, simplicity, and an abundant literature com-These data may reflect the presence of low-grade in-piled over the past seven decades. Nonetheless,flammation in coronary arteries or elsewhere, or al-measurement of C-reactive protein has several ad-ternatively, they may reflect proinflammatory or pro-vantages over this method. The erythrocyte sedi-thrombotic effects of C-reactive protein itself.41,42mentation rate is an indirect measurement of plas-These findings are not likely to prove clinically useful,ma acute-phase protein concentrations and can besince the mildly elevated concentrations in thesegreatly influenced by the size, shape, and number ofstudies fall well within the range in healthy subjects.erythrocytes, as well as by other plasma constituentsPlasma concentrations of cytokines and cytokinesuch as immunoglobulins. Consequently, the resultsreceptors have been studied in patients with inflam-are imprecise and sometimes misleading. Althoughmatory conditions. Measurement of cytokines in plas-the erythrocyte sedimentation rate represented ama is difficult, partly because of their short plasmagreat advance when it was introduced in the 1920s,half-lives and the presence of blocking factors.61,62this indirect method (which some regard as archaic)Plasma interleukin-6 concentrations are elevated inis no longer needed to assess plasma concentrationspatients with many inflammatory diseases, but ex-of fibrinogen, because they can now be determinedcept for the rapidity with which change occurs, meas-directly. As a patient’s condition worsens or im-urement of plasma interleukin-6 concentrations hasproves, the erythrocyte sedimentation rate changesno apparent advantage over measurement of plasmarelatively slowly, whereas plasma C-reactive proteinC-reactive protein. Reports of different patterns ofconcentrations change rapidly. The range of abnor-cytokine responses in different disease states raisemal values for C-reactive protein is broader than thethe possibility that cytokine determinations may ul-range of abnormal values for the erythrocyte sedi-timately have diagnostic value.63,64 Until further stud-mentation rate, with accompanying clinical implica-ies are available, the high cost, limited availability,tions: among patients with plasma C-reactive pro-and absence of standardization argue against thetein concentrations greater than 100 mg per liter, 80measurement of plasma cytokines and their recep-to 85 percent have bacterial infections.4 The eryth-tors in clinical practice.452·Fe b r u a r y 11 , 19 9 9Downloaded from www.nejm.org by VICTOR ARREDONDO MD on March 27, 2007 . Copyright © 1999 Massachusetts Medical Society. All rights reserved. M EC H A N I S M S O F D I S E AS ECONCLUSIONS14. Xing Z, Gauldie J, Cox G, et al. IL-6 is an antiinflammatory cytokine required for controlling local or systemic acute inflammatory responses. The acute-phase response, an important pathophys-J Clin Invest 1998;101:311-20.iologic phenomenon, replaces the normal homeo-15. Mackiewicz A, Speroff T, Ganapathi MK, Kushner I. Effects of cyto-kine combinations on acute phase protein production in two human static mechanisms with new set points that presum-hepatoma cell lines. J Immunol 1991;146:3032-7.ably contribute to defensive or adaptive capabilities.16. Gabay C, Smith MF, Eidlen D, Arend WP. Interleukin 1 receptor an-The functions of these changes are highly variabletagonist (IL-1Ra) is an acute-phase protein. J Clin Invest 1997;99:2930-40.and diverse: some participate in initiating or sustain-17. Loyer P, Ilyin G, Abdel Razzak Z, et al. Interleukin 4 inhibits the pro-ing the inflammatory process, others modulate it,duction of some acute-phase proteins by human hepatocytes in primary and still others have adaptive roles. These changesculture. FEBS Lett 1993;336:215-20.18. Mackiewicz A, Shooltink H, Heinrich PC, Rose-John S. Complex of are induced by a complex intercellular signaling sys-soluble human IL-6-receptor/IL-6 up-regulates expression of acute-phase tem of which the chief constituents are inflam-proteins. J Immunol 1992;149:2021-7.19. Baumann H, Richard C, Gauldie J. Interaction among hepatocyte-mation-associated cytokines. Several cytokines, par-stimulating factors, interleukin 1, and glucocorticoids for regulation of ticularly interleukin-6, stimulate the production ofacute phase plasma proteins in human hepatoma (HepG2) cells. J Immunol acute-phase proteins in response to varied stimuli.1987;139:4122-8.20. Campos SP, Wang Y, Koj A, Baumann H. Insulin cooperates with IL-1 The patterns of cytokine production and of thein regulating expression of a1-acid glycoprotein gene in rat hepatoma cells. acute-phase response differ in different inflammato-Cytokine 1994;6:485-92.ry conditions. Acute-phase changes reflect the pres-21. Jiang S-L, Samols C, Rzewnicki D, et al. Kinetic modeling and math-ematical analysis indicate that acute phase gene expression in Hep 3B cells ence and intensity of inflammation, and they haveis regulated by both transcriptional and posttranscriptional mechanisms. long been used as a clinical guide to diagnosis andJ Clin Invest 1995;95:1253-61.22. Rogers JT, Bridges KR , Durmowicz GP, Glass J, Auron PE, Munro management. For this purpose, determination of se-HN. Translational control during the acute phase response: ferritin synthe-rum C-reactive protein has advantages over the tra-sis in response to interleukin-1. J Biol Chem 1990;265:14572-8.ditional strategy of measuring the erythrocyte sedi-23. Van Dijk W, Mackiewicz A. Interleukin-6-type cytokine-induced changes in acute phase protein glycosylation. Ann N Y Acad Sci 1995;762:mentation rate.319-30.24. de Graaf TW, Van der Stelt ME, Anbergen MG, van Dijk W. Inflam-mation-induced expression of sialyl Lewis X-containing glycan structures Supported by a grant (AG 02467) from the National Institutes of Healthon alpha 1-acid glycoprotein (orosomucoid) in human sera. J Exp Med (to Dr. Kushner) and a postdoctoral fellowship grant from the Swiss Na-1993;177:657-66.tional Foundation for Scientific Research and the Fondation Suisse de25. Dube R , Rook GAW, Steele J, et al. Agalactosyl IgG in inflammatory Bourse de Médecine et Biologie (to Dr. Gabay).bowel disease: correlation with C-reactive protein. Gut 1990;31:431-4.26. Yue CC, Muller-Greven J, Dailey P, Lozanski G, Anderson V, Macin-We are indebted to Drs. John W. Harris, William P. Arend, Bar-tyre S. Identification of a C-reactive protein binding site in two hepatic car-bara Barna, M. Asim Khan, and Mojtaba Youssefi for critical re-boxylesterases capable of retaining C-reactive protein within the endoplas-views of the manuscript, to Dr. Carl Grunfeld for useful conversa-mic reticulum. J Biol Chem 1996;271:22245-50.27. Dinarello CA. Cytokines as endogenous pyrogens. In: Mackowiak PA, tions, and to Ms. Debra Rzewnicki for editorial assistance.ed. Fever: basic mechanisms and management. 2nd ed. Philadelphia: Lip-pincott-Raven, 1997:87-116.REFERENCES28. Goldbach JM, Roth J, Zeisberger E. Fever suppression by subdia-phragmatic vagotomy in guinea pigs depends on the route of pyrogen ad-1. Tillett WS, Francis T Jr. Serological reactions in pneumonia with non-ministration. Am J Physiol 1997;272:R675-R681.protein somatic fraction of pneumococcus. J Exp Med 1930;52:561-71.29. Chrousos GP. The hypothalamic-pituitary-adrenal axis and immune-2. Kushner I. Regulation of the acute phase response by cytokines. Per-mediated inflammation. N Engl J Med 1995;332:1351-62.spect Biol Med 1993;36:611-22.30. Sarraf P, Frederich RC, Turner EM, et al. Multiple cytokines and acute 3. Dinarello CA. Interleukin-1 and the pathogenesis of the acute-phase re-inflammation raise mouse leptin levels: potential role in inflammatory an-sponse. N Engl J Med 1984;311:1413-8.orexia. J Exp Med 1997;185:171-5.4. Morley JJ, Kushner I. Serum C-reactive protein levels in disease. Ann 31. Means RT Jr. Pathogenesis of the anemia of chronic disease: a cyto-N Y Acad Sci 1982;389:406-18.kine-mediated anemia. Stem Cells 1995;13:32-7.5. Gitlin JD, Colten HR. Molecular biology of the acute phase plasma 32. Weiss G, Bogdan C, Hentze MW. Pathways for the regulation of mac-proteins. In: Pick E, Landy M, eds. Lymphokines. Vol. 14. San Diego, Cal-rophage iron metabolism by the anti-inflammatory cytokines IL-4 and if.: Academic Press, 1987:123-53.IL-13. J Immunol 1997;158:420-5.6. Maes M, Delange J, Ranjan R , et al. Acute phase proteins in schizo-33. Zeidler C, Kanz L, Hurkuck KL, et al. In vivo effects of interleukin-phrenia, mania and major depression: modulation by psychotropic drugs. 6 on thrombopoiesis in healthy and irradiated primates. Blood 1992;80:Psychiatry Res 1997;66:1-11.2740-5.7. Wigmore SJ, Fearon KCH, Maingay JP, Lai PBS, Ross JA. Interleukin-34. Tai H, Miyaura C, Pilbeam CC, et al. Transcriptional induction of cy-8 can mediate acute-phase protein production by isolated human hepato-clooxygenase-2 in osteoblasts is involved in interleukin-6-induced osteo-cytes. Am J Physiol 1997;273:E720-E726.clast formation. Endocrinology 1997;138:2372-9.8. Gauldie J, Richards C, Harnish D, Lansdorp P, Baumann H. Interferon 35. Moldawer LL, Copeland EM III. Proinflammatory cytokines, nutri-b2/B-cell stimulatory factor type 2 shares identity with monocyte-derived tional support, and the cachexia syndrome: interactions and therapeutic hepatocyte-stimulating factor and regulates the major acute phase protein options. Cancer 1997;79:1828-39.response in liver cells. Proc Natl Acad Sci U S A 1987;84:7251-5.36. Hardardottir I, Grunfeld C, Feingold KR. Effects of endotoxin and 9. Fattori E, Cappelletti M, Costa P, et al. Defective inflammatory re-cytokines on lipid metabolism. Curr Opin Lipidol 1994;5:207-15.sponse in interleukin 6-deficient mice. J Exp Med 1994;180:1243-50.37. Wolf M, Bohm S, Brand M, Kreymann G. Proinflammatory cytokines 10. Zheng H, Fletcher D, Kozak W, et al. Resistance to fever induction interleukin 1 beta and tumor necrosis factor a inhibit growth hormone and impaired acute-phase response in interleukin-1 b-deficient mice. Im-stimulation of insulin-like growth factor I synthesis and growth hormone munity 1995;3:9-19.receptor mRNA levels in cultured rat liver cells. Eur J Endocrinol 1996;11. Richards CD, Langdon C, Pennica D, Gauldie J. Murine cardiotro-135:729-37.phin-1 stimulates the acute-phase response in rat hepatocytes and H35 38. De Benedetti F, Alonzi T, Moretta A, et al. Interleukin 6 causes hepatoma cells. J Interferon Cytokine Res 1996;16:69-75.growth impairment in transgenic mice through a decrease in insulin-like 12. Feldmann M, Brennan FM, Maini RN. Role of cytokines in rheuma-growth factor-1: a model for stunted growth in children with chronic in-toid arthritis. Annu Rev Immunol 1996;14:397-440.flammation. J Clin Invest 1997;99:643-50.13. Dinarello CA. Biologic basis for interleukin-1 in disease. Blood 1996;39. Kushner I. Semantics, inflammation, cytokines and common sense. 87:2095-147.Cytokine Growth Factor Rev 1998;9:191-6.Vo l u m e 3 4 0Nu m b e r 6·453Downloaded from www.nejm.org by VICTOR ARREDONDO MD on March 27, 2007 . Copyright © 1999 Massachusetts Medical Society. All rights reserved. The Ne w E n g l a nd J o u r n a l o f Me d ic i ne40. Volanakis JE. Acute phase proteins in rheumatic disease. In: Koopman 53. Bone RC. Toward a theory regarding the pathogenesis of the systemic WJ, ed. Arthritis and allied conditions: a textbook of rheumatology. 13th inflammatory response syndrome: what we do and do not know about cy-ed. Baltimore: Williams & Wilkins, 1997:505-14.tokine regulation. Crit Care Med 1996;24:163-72.41. Ballou SP, Lozanski G. Induction of inflammatory cytokine release 54. van Leeuwen MA, van der Heijde DMFM, van Rijswijk MH, et al. from cultured human monocytes by C-reactive protein. Cytokine 1992;Interrelationship of outcome measures and process variables in early rheu-4:361-8.matoid arthritis: a comparison of radiologic damage, physical disability, 42. Cermak J, Key NS, Bach RR , Balla J, Jacob HS, Vercellotti GM. joint counts, and acute phase reactants. J Rheumatol 1994;21:425-9.C-reactive protein induces human peripheral blood monocytes to synthe-55. Pepys MB, Lanham JG, De Beer FC. C-reactive protein in SLE. Clin size tissue factor. Blood 1993;82:513-20.Rheum Dis 1982;8:91-103.43. Ahmed N, Thorley R , Xia D, Samols D, Webster RO. Transgenic mice 56. ter Borg EJ, Horst G, Limburg PC, van Rijswijk MH, Kallenberg expressing rabbit C-reactive protein exhibit diminished chemotactic factor-CGM. C-reactive protein levels during disease exacerbations and infections induced alveolitis. Am J Respir Crit Care Med 1996;153:1141-7.in systemic lupus erythematosus: a prospective longitudinal study. J Rheu-44. Xia D, Samols D. Transgenic mice expressing rabbit C-reactive protein matol 1990;17:1642-8.are resistant to endotoxemia. Proc Natl Acad Sci U S A 1997;94:2575-57. Moutsopoulos HM, Mavridis AK, Acritidis NC, Avgerinos PC. High 80.C-reactive protein response in lupus polyarthritis. Clin Exp Rheumatol 45. Zouki C, Beauchamp M, Baron C, Filep JG. Prevention of in vitro 1983;1:53-5.neutrophil adhesion to endothelial cells through shedding of L-selectin by 58. Spector TD, Hart DJ, Nandra D, et al. Low-level increases in serum C-reactive protein and peptides derived from C-reactive protein. J Clin In-C-reactive protein are present in early osteoarthritis of the knee and predict vest 1997;100:522-9.progressive disease. Arthritis Rheum 1997;40:723-7.46. Malle E, De Beer FC. Human serum amyloid A (SAA) protein: a 59. Haverkate F, Thompson SG, Pyke SDM, Gallimore JR , Pepys MB. prominent acute-phase reactant for clinical practice. Eur J Clin Invest 1996;Production of C-reactive protein and risk of coronary events in stable and 26:427-35.unstable angina. Lancet 1997;349:462-6.47. Banka CL, Yuan T, de Beer MC, Kindy M, Curtiss LK, de Beer FC. 60. Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH. In-Serum amyloid A (SAA): influence on HDL-mediated cellular cholesterol flammation, aspirin, and the risk of cardiovascular disease in apparently efflux. J Lipid Res 1995;36:1058-65.healthy men. N Engl J Med 1997;336:973-9. [Erratum, N Engl J Med 48. Berliner JA, Navab M, Fogelman AM, et al. Atherosclerosis: basic 1997;337:356.]mechanism — oxidation, inflammation, and genetics. Circulation 1995;91:61. May LT, Viguet H, Kenney JS, Ida N, Allison AC, Sehgal PB. High 2488-96.levels of “complexed” interleukin-6 in human blood. J Biol Chem 1992;49. Kilpatrick L, McCawley L, Nachiappan V, et al. a-1-Antichymotrypsin 267:19698-704.inhibits the NADPH oxidase-enzyme complex in phorbol ester-stimulated 62. Arend WP, Malyak M, Smith MF Jr, et al. Binding of IL-1a, IL-1b, neutrophil membranes. J Immunol 1992;149:3059-65.and IL-1 receptor antagonist by soluble IL-1 receptors and levels of soluble 50. Cid MC, Grant DS, Hoffman GS, Auerbach R , Fauci AS, Kleinman IL-1 receptors in synovial fluids. J Immunol 1994;153:4766-74.HK. Identification of haptoglobin as an angiogenic factor in sera from pa-63. Gabay C, Cakir N, Moral F, et al. Circulating levels of tumor necrosis tients with systemic vasculitis. J Clin Invest 1993;91:977-85.factor soluble receptors in systemic lupus erythematosus are significantly 51. Borish L, King MS, Mascali JJ, Johnson S, Coll B, Rosenwasser LJ. higher than in other rheumatic diseases and correlate with disease activity. Transthyretin is an inhibitor of monocyte and endothelial cell interleukin-J Rheumatol 1997;24:303-8.1 production. Inflammation 1992;16:471-84.64. Gabay C, Gay-Croisier F, Roux-Lombard P, et al. Elevated serum lev-52. Kluger MJ, Kozak W, Conn CA, Leon LR , Soszynski D. The adaptive els of interleukin-1 receptor antagonist in polymyositis/dermatomyositis: a value of fever. In: Mackowiak PA, ed. Fever: basic mechanisms and man-biologic marker of disease activity with a possible role in the lack of acute-agement. 2nd ed. Philadelphia: Lippincott-Raven, 1997:255-66.phase protein response. Arthritis Rheum 1994;37:1744-51.454·Fe b r u a r y 11 , 19 9 9Downloaded from www.nejm.org by VICTOR ARREDONDO MD on March 27, 2007 . Copyright © 1999 Massachusetts Medical Society. All rights reserved. New England Journal of MedicineCORRECTIONAcute-Phase Proteins and Other SystemicResponses to InﬂammationAcute-Phase Proteins and Other Systemic Responses to Inﬂamma-tion . On page 449, in Table 1, under the heading ``Proteins whoseplasma concentrations decrease,´´ retinol-binding protein should haveappeared. Also on page 449, in Table 2, the last entry should havebeen two entries reading, ``Decreased plasma retinol concentrations´´and ``Increased plasma glutathione concentrations,´´ not one entryreading, ``Increased plasma retinol and glutathione concentrations,´´as printed.N Engl J Med 1999;340:1376Downloaded from www.nejm.org by VICTOR ARREDONDO MD on March 27, 2007 . Copyright © 1999 Massachusetts Medical Society. All rights reserved.