REVIEW Cold antibody autoimmune hemolytic anemias Lawrence D. Petz * Emeritus Professor of Pathology and Laboratory Medicine, University of California Los Angeles, StemCyte International Cord Blood Center, Arcadia, California, United States Summary The cold antibody autoimmune hemolytic anemias (AIHAs) are primarily comprised of cold agglutinin syndrome (CAS) and paroxysmal cold hemoglobinuria (PCH) but, in addition, there are unusual instances in which patients satisfy the sero- logic criteria of both warm antibody AIHA and CAS (‘‘mixed AIHA’’). CAS characteristi- cally occurs in middle-aged or elderly persons, often with signs and symptoms exacerbated by cold. The responsible antibody is of the IgM immunoglobulin class, is maximally reactive in the cold but with reactivity up to at least 30 °C. Therapy is often ineffective,butneweragentssuchasrituximabhavebeenbeneficialinsomepatients. PCH occurs primarily in children, often after an upper respiratory infection. The caus- ative antibody is of the IgG immunoglobulin class and is a biphasic hemolysin that is demonstratedbyincubationinthecoldfollowedbyincubationat37 °Cinthepresence of complement. Acute attacks are frequently severe but the illness characteristically resolvesspontaneouslywithinafewdaystoseveralweeksafteronsetandrarelyrecurs. Treatment consists of supportive care, with transfusions frequently being needed. c 2007 Elsevier Ltd. All rights reserved. KEYWORDS Autoimmune hemolytic anemia; Cold agglutinin syndrome; Paroxysmal cold hemoglobinuria Introduction The immune hemolytic anemias are classified as indicated in Table 1. 1 This classification divides these disorders into distinctive categories which have differing clinical manifestations, prognosis and therapy, as indicated in Table 2. The cold anti- body autoimmune hemolytic anemias (AIHAs) are primarily comprised of cold agglutinin syndrome (CAS) and paroxysmal cold hemoglobinuria (PCH) but,inaddition,thereareunusualinstancesinwhich patients satisfy the serologic criteria of both warm antibody AIHA and CAS. Such cases are designated as combined cold and warm AIHA or ‘‘mixed AIHA.’’ These disorders are diagnosed on the basis of characteristic serologic reactions. The majority of cases of AIHA are mediated by warm-reactive auto- antibodies, i.e., antibodies displaying optimal reactivity with human RBC at 37 °C and which are usually of the IgG immunoglobulin class. In con- trast, CAS is generally caused by IgM autoantibod- ies which exhibit maximal reactivity at 4 °C. The causative in PCH antibody is an IgG immuno- globulin with specificity that differs from that found in CAS. The antibody is best detected in vitro by its ability to cause hemolysis of normal RBC in a two- step procedure, which requires incubation in the 0268-960X/$ - see front matter c 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.blre.2007.08.002 * Tel.: +1 626 821 9860; fax: +1 626 821 9660. E-mail address: [email protected]Blood Reviews (2008) 22, 1–15 www.elsevierhealth.com/journals/blre
Transcript
Blood Reviews (2008) 22, 1–15
www.elsevierhealth.com/journals/blre
REVIEW
Cold antibody autoimmune hemolytic anemias
Lawrence D. Petz *
Emeritus Professor of Pathology and Laboratory Medicine, University of California Los Angeles,StemCyte International Cord Blood Center, Arcadia, California, United States
Summary The cold antibody autoimmune hemolytic anemias (AIHAs) are primarilycomprised of cold agglutinin syndrome (CAS) and paroxysmal cold hemoglobinuria(PCH) but, in addition, there are unusual instances in which patients satisfy the sero-logic criteria of both warm antibody AIHA and CAS (‘‘mixed AIHA’’). CAS characteristi-cally occurs in middle-aged or elderly persons, often with signs and symptomsexacerbated by cold. The responsible antibody is of the IgM immunoglobulin class, ismaximally reactive in the cold but with reactivity up to at least 30 �C. Therapy is oftenineffective, but newer agents such as rituximab have been beneficial in some patients.PCH occurs primarily in children, often after an upper respiratory infection. The caus-ative antibody is of the IgG immunoglobulin class and is a biphasic hemolysin that isdemonstrated by incubation in the cold followed by incubation at 37 �C in the presenceof complement. Acute attacks are frequently severe but the illness characteristicallyresolves spontaneouslywithin a fewdays to severalweeksafter onsetand rarely recurs.Treatment consists of supportive care, with transfusions frequently being needed.
The immune hemolytic anemias are classified asindicated in Table 1.1 This classification dividesthese disorders into distinctive categories whichhave differing clinical manifestations, prognosisand therapy, as indicated in Table 2. The cold anti-body autoimmune hemolytic anemias (AIHAs) areprimarily comprised of cold agglutinin syndrome(CAS) and paroxysmal cold hemoglobinuria (PCH)but, in addition, there are unusual instances inwhichpatients satisfy the serologic criteria of both warm
268-960X/$ - see front matter �c 2007 Elsevier Ltd. All rights reseoi:10.1016/j.blre.2007.08.002
antibody AIHA and CAS. Such cases are designatedas combined cold and warm AIHA or ‘‘mixed AIHA.’’
These disorders are diagnosed on the basis ofcharacteristic serologic reactions. The majority ofcases of AIHA are mediated by warm-reactive auto-antibodies, i.e., antibodies displaying optimalreactivity with human RBC at 37 �C and which areusually of the IgG immunoglobulin class. In con-trast, CAS is generally caused by IgM autoantibod-ies which exhibit maximal reactivity at 4 �C.
The causative in PCH antibody is an IgG immuno-globulin with specificity that differs from that foundin CAS. The antibody is best detected in vitro by itsability to cause hemolysis of normal RBC in a two-step procedure, which requires incubation in the
Alloantibody-induced immune hemolytic anemiaHemolytic transfusion reactionsHemolytic disease of the fetus and newborn
Table 2 Some characteristic features of autoimmune and drug induced immune hemolytic anemias.
Warm antibody AIHAClinical manifestations: Variable, usually symptoms of anemia, occasionallyAcute hemolytic syndromePrognosis: Fair, with significant mortalityMost effective therapies: Steroids, splenectorny, immunosuppressive drugs
Cold agglutinin syndromeClinical manifestations: Moderate chronic hemolytic anemia in middle-aged or elderly person, often withsigns and symptoms exacerbated by coldPrognosis: Good, usually a chronic and quite stable anemiaMost effective therapies: Avoidance of cold exposure, immunosuppressive drugs
Paroxysmal cold hemoglobinuriaClinical manifestations: Acute hemolytic anemia, often with hemoglobinuria, particularly in a child withhistory of recent viral or viral-like illnessPrognosis: Excellent after initial stormy courseTherapy: Not well defined; steroids empirically and transfusions if required
Drug-induced immune hemolytic anemiaClinical manifestations: Variable, most commonly subacute in onset, but occasionally acute hemolytic syndromePrognosis: ExcellentTherapy: Stop drug; occasionally a short course of steroids empirically
2 L.D. Petz
cold followed by incubation at 37 �C in the presenceof complement.
AIHA is designated as ‘‘idiopathic’’ if it is unas-sociated with any demonstrable underlying dis-ease. In contrast, AIHA is categorized as‘‘secondary’’ if it is associated with an additionaldisorder and there is reason to suspect that theassociation is not merely fortuitous.
The ingestion of some drugs causes hemolyticanemia in which the causative antibody can beshown to have specificity for the drug or its metabo-
lites. Although the antibody reacts with thepatient’s own RBCs, these disorders are not auto-immune disorders because the antibody does nothave specificity for autoantigens. In other quiteremarkable cases, ingestion of a drug causes thedevelopment of red cell autoantibodies, i.e., theantibody in the patient’s serumand in an eluate fromthe patient’s red cells reacts with red cells similarlyto autoantibodies in idiopathic AIHAs, and no rela-tionship between the drug and the antibody can bedemonstrated in vitro. Such cases are appropriately
Cold antibody autoimmune hemolytic anemias 3
termed drug-induced AIHAs and can be either ofwarm antibody or cold antibody type.
Cold agglutinin syndrome
Incidence
Cold agglutinin syndrome (CAS) is relatively uncom-mon compared with warm antibody AIHA (WAIHA),but it occurs much more frequently than PCH. In aseries of 347 patients with acquired immune hemo-lytic anemias (including drug-induced immunehemolytic anemia), 54 patients (15.6%) had CAS.2
Dacie and Worlledge3 reported that 25 percent oftheir patients with AIHA had CAS, whereas van Log-hem et al4 and Dausset and Colombani5 reported anincidence of only 7.7%.
Age and sex
Patients with idiopathic CAS are typically middle-aged or elderly. Worlledge6 reported that the aver-age age of 29 patients was 66. In Dacie’s series,only 3 of 21 patients with idiopathic CAS wereyounger than 50, whereas 9 of 12 post-pneumoniacases occurred in patients between the ages of 20and 45.7 Chronic idiopathic cold agglutinin diseaseoccurs only very rarely in childhood.7–9
Clinical characteristics
Characteristic Clinical Manifestations are indicatedin Table 3.
Dark Urine (especially in cold weather)Acrocyanosis (occasionally leading to gangrene of digits)Pallor and JaundiceHepatosplenomegaly and Lymphadenopathy may be pres
Laboratory FindingsMild to Moderate Anemia
Prominent autoagglutination, especially at cold tempAbnormal RBC Morphology
Modest degrees of Spherocytosis, Anisocytosis, PoikiloPolychromatophilia
ReticulocytosisJaundiceHemoglobinuriaErythroid Hyperplasia in the bone marrow
SymptomsThe clinical manifestations vary greatly frompatient to patient, probably depending on the ther-mal range of the cold antibody. Symptoms are fre-quently just those of a chronic anemia whichprogresses gradually.10 In idiopathic cases, the dis-order often remains quite static and generally onlyprogresses slowly in intensity. Clinical findings aremore serious for patients whose cold agglutinin isactive at higher temperatures.11 Temperatures of30 �C and lower are normally attained in the super-ficial skin vessels of those parts of the bodyexposed to cold air or water.12 The thermal rangeof the antibody is more important than the aggluti-nation titer for clinical purposes.11,13
HemoglobinuriaPatients may experience hemoglobinuria, particu-larly in cold weather, and this finding is prominentamong manifestations listed in reviews of the disor-der. However, its incidence is difficult to deter-mine, and in many patients hemoglobinuria isnever observed. This is true even in some patientswho develop severe acrocyanosis during cold expo-sure.7,14When an episode of intravascular hemolysisdoes occur after cold exposure, it is not associatedwith fever, chills, or acute renal insufficiency.12,15
This phenomenon contrasts with the marked consti-tutional symptoms associated with hemoglobinuriain PCH or in those few patients with warm antibodyAIHA who have hemoglobinuria.
AcrocyanosisMany affected patients complain of acrocyanosis ofthe ears, the nose tip, fingers, and toes in cold
Manifestations.
ent.
eratures.
cytosis,
4 L.D. Petz
temperatures, a condition that vanishes quickly onwarming.11,16 These symptoms are caused by auto-agglutination of the patient’s erythrocytes in theskin capillaries causing local blood stasis.
GangreneOccasionally actual gangrene, particularly of a digitor digits, hasbeenobserved.17,18Although theoccur-rence of gangrene after prolonged exposure to coldwould appear to be a logical consequence of vascularinsufficiency causedby autohemagglutination, thereis some question as to whether the presence of cryo-globulins may not be an additive factor.
Raynaud’s phenomenonThe acrocyanosis associated with cold agglutininsmay be distinguished from the much more commonRaynaud’s phenomenon that may be idiopathic orassociated with diseases such as collagen vasculardisorders.10 In the former, the exposed distal fin-gers or toes turn a dusky blue that either quicklyreverts to normal on warming or may progress toblanching after more prolonged or more severe coldexposure. In Raynaud’s phenomenon, the digitscharacteristically turn white from vasospasm andthen, as ischemia is resolving, turn blue; thischange may be followed by a hyperemic phase inwhich the exposed parts become reddened. Fur-ther, exposure to moderate degrees of cold inpatients with cold agglutinins will cause all digitsto be equally affected, whereas Raynaud’s pheno-menon may cause marked blanching of one or twofingers with the others remaining normal.
Other physical findingsPallor and jaundice may be present and the inten-sity of these findings depends on the rate of hemo-lysis and the ability of the liver to excrete bilirubin.Hepatosplenomegaly and lymphadenopathy are notprominent findings.11 Dacie7 indicates that thespleen is palpably enlarged only in a minority ofpatients, whereas Schubothe12 states that usuallythe spleen and sometimes the liver are slightlyenlarged. Cutaneous necrosis is an unusual mani-festation of cold agglutinin syndrome.19–22
Laboratory findings
AutoagglutinationAutoagglutination of anticoagulated blood samplesthat occurs quickly as blood cools to room temper-ature is characteristic and is frequently the firstobservation made to suggest the diagnosis. Auto-agglutination is intensified by cooling the blood to4 �C and is reversed by warming to 37 �C. This prop-
erty distinguishes cold autoagglutination from rou-leaux formation and also from autoagglutinationwhich occurs in some patients with warm antibodyAIHA. In the latter case, the autoagglutination doesnot resolve at 37 �C. The agglutination createsproblems in making satisfactory blood films andperforming blood counts.11,23–26 However, it shouldbe stressed that clinically insignificant cold autoan-tibodies will also cause autoagglutination. Suchantibodies have a low thermal amplitude and donot cause hemolysis.27 Therefore, the mere obser-vance of cold autoagglutination is not diagnosticof cold agglutinin disease, and the cold antibodiesmust be further characterized as described below.A rather common diagnostic error is over-diagnosisof CAS in patients who have benign cold antibodiesand pathogenic warm autoantibodies.
Peripheral bloodRBC morphology is less abnormal than in patientswith warm antibody AIHA, with lesser degrees ofanisocytosis and poikilocytosis.7 Similarly, sphero-cytosis is less intense but has been noted by Leddyand Swisher,28 Schreiber et al29 and Rosse.30 Leu-kocyte and platelet counts are usually normal.Bilirubinemia of mild degree is common. Reticulo-cytosis usually is in proportion to the severity ofhemolysis.
Bone marrowBone marrow examination shows a variabledegree of erythroid hyperplasia and a slight butdefinite increase of lymphoid cells.11 Severalauthors have reported lymphoid proliferation insome cases which may become progressivelymore extensive, especially in the terminal stagesof the disease.12,31,32
Diagnostic Serologic Tests for CAS. The diagno-sis must be considered in all patients with acquiredhemolytic anemia who have a positive direct anti-globulin test (DAT) using anti-C3 and a negativeDAT using anti-IgG. A practical initial serum screen-ing procedure is to test the ability of the patient’sserum to agglutinate saline-suspended normal redcells at 20 �C (or room temperature) after incuba-tion for 30–60 minutes. This recommendation isbased on the fact that the sera of essentially allpatients series who have CAS cause agglutinationof saline-suspended red cells at 20 �C. If thisscreening test is negative, cold agglutinin syn-drome is extremely unlikely; if positive, furtherstudies are necessary to determine the thermalamplitude of the antibody.
When CAS appears to be a possible diagnosis onthe basis of the preceding evaluation, studies ofthe thermal range of reactivity of the antibody in
Cold antibody autoimmune hemolytic anemias 5
saline and albumin are indicated. It is also conve-nient to determine possible li blood group specifi-city of the antibody simultaneously. The titer ofthe cold agglutinin in CAS is invariably highest at4 �C and progressively decreases at higher temper-atures. Of particular note are the reactions at 30�C. If the reaction is positive in saline or albumin,the antibody may well be of pathogenetic signifi-cance, i.e., it may be causing short red blood cellsurvival in vivo.
Utilizing clinical information, the results of theDAT, and the preceding screening tests, a reason-ably confident assessment of the presence or ab-sence of CAS may be made. CAS may bediagnosed if the following are present: (1) Clini-cal evidence of acquired hemolytic anemia; (2)a positive DAT caused by sensitization with C3;(3) a negative DAT using anti-IgG; (4) the pres-ence of a cold autoagglutinin with reactivity upto at least 30 �C in saline or albumin; and (5) acold agglutinin titer (at 4 �C) P 256, except inexceptional cases.
An alternative diagnosis must be sought for pa-tients who do not satisfy all these criteria.
Course and prognosis
The prognosis in patients with idiopathic CAS issignificantly better than it is in patients withwarm antibody AIHA. Most commonly, the patienthas a chronic, fairly stable anemia that is mild ormoderate in severity. For example, in only 2 ofthe 12 patients reviewed by Ferriman et al14
did the hemoglobin fall below 7.4 gm/dl. Simi-larly, Schubothe12 describes the course of thedisease as ‘‘monotonous,’’ and patients may sur-vive many years, suffering minimal disability. Hereported 16 patients followed for ‘‘severalyears,’’ and every patient had a hemoglobin of9 mg/dl or greater at some time during the per-iod, and in only 2 patients did the hemoglobinfall below 6.5 gm/dI at any time. This findingcontrasts strikingly with hemoglobin values inwarm AIHA or PCH.
Most patients tolerate their mild or moderateanemia quite well. In more severe cases, however,death may ensue from complications of slowly pro-gressive anemia, or from complications of bloodtransfusion. Sometimes, after a considerable time,a generalized malignant lymphoproliferation mayoccur which is manifested in the bone marrow mor-phology, as described above. Occasionally, anintensification of hemolysis with an increase ofthe thermal amplitude up to the range of the inter-nal body temperature seems to lead to a fatalcomplication.
Management
Avoidance of coldTherapy for cold agglutinin syndrome is often unsat-isfactory. It is fortunate that the disorder is fre-quently not severe and, instead, results in achronic mild hemolytic anemia with a hemoglobinlevel of 9–12 gm per dL. Such patients require notherapy other than avoidance of cold.33 If patientsgo out in cold weather, theymust clothe themselveswell, including earmuffs, warm socks, and warmmittens. These measures will prevent the develop-ment of acute hemolytic crises but even with strictavoidance of cold, some hemolysis usually persists.Patients with moderate degrees of anemia shouldusually be urged to learn to tolerate the symptomsrather than embark on therapeutic trials that havegreater potential risk than probable benefit.
Corticosteroid therapyCorticosteroids are less effective in cold agglutininsyndrome than in warm antibody syndromes.33 Lackof success of corticosteroid therapy in cold aggluti-nin syndrome has been reported by Pisciotta,34
Firkin et al,35 Dausset and Colombani,5 and Dacie.33
However, some case reports do indicate that corti-costeroids can be effective, often in patients withatypical clinical and serological findings.29,36,37
Immunosuppressive drugsCyclosphosphamide and chlorambucil haveresulted in favorable responses in a minority ofpatients with CAS.12,38,39 In patients who do notrespond satisfactorily to therapy with immunosup-pressive drugs, bone marrow depression and tem-porary worsening of anemia result. Therefore, itis wise to begin with low doses orally, e.g., dailydoses of chlorambucil of 2 or 4 mg per day. Bi-weekly blood counts including a reticulocyte countshould be obtained, and the daily dose may be in-creased by 2 mg every 2 months until favorableeffects or limiting adverse reactions result. Thelatter effects usually consist of marrow suppres-sion, although anorexia and nausea may limit dos-age as well. If a favorable response does result,maintenance therapy may be advisable, althoughan alternative approach is to use therapy intermit-tently as needed.38,40
FludarabineJacobs41 described a patient with idiopathic coldagglutinin disease who initially responded tocyclophosphamide-containing chemotherapy butsubsequentlybecamerefractory to this agent.Treat-ment was changed to fludarabine, 25 mg/mm2 ofbody surface area daily for 5 days intravenously.
6 L.D. Petz
The treatmentwas repeated every 28 days for a totalof three courses. This resulted in a clinical remissionwith a normal hemoglobin, reticulocyte count andbilirubin that had lasted 4 years by the time of publi-cation. Although fludarabine therapy has beenknown to cause AIHA,42 Jacobs41 nevertheless sug-gested that consideration should be given to fludara-bine’s use in patients with idiopathic CAS who arerefractory to conventional alkylating agents.
RituximabA number of reports indicate that Rituximab43 is aneffective drug in some cases of CAS.44–47 Berentsenet al48 reported on 37 courses of rituximab admin-istered prospectively to 27 patients with primarychronic CAS. Fourteen of 27 patients respondedto their first course of rituximab, and 6 of 10responded to re-treatment. In both groups com-bined, responses were achieved after 20 of 37courses, giving an overall response rate of 54%.The investigators observed 1 complete and 19 par-tial responses. Two nonresponders and 3 patientswho experienced relapse received second-linetherapy with interferon-alpha combined with anew course of rituximab, and 1 nonresponder and2 patients who experienced relapse achieved par-tial responses. Responders achieved a medianincrease in hemoglobin levels of 40 g/L (4 g/dL).Median time to response was 1.5 months, and med-ian observed response duration was 11 months. Theauthors concluded that rituximab is an effectiveand well-tolerated therapy for CAS.
Schollkopf et al49 performed a phase II multi-center trial to investigate the effect of rituximabin CAS, including 20 patients studied from October2002 until April 2003. Thirteen patients had idio-pathic CAS and seven patients had CAS associatedwith a malignant B-cell lymphoproliferative dis-ease. Rituximab was given in doses of 375 mg/m2
at days 1, 8, 15 and 22. Sixteen patients were fol-lowed up for at least 48 weeks. Four patients wereexcluded after 8, 16, 23 and 28 weeks for reasonsunrelated to CAS. Nine patients (45%) respondedto the treatment, one with complete response,and eight with partial response. Eight patients re-lapsed, one patient was still in remission at theend of follow-up. There were no serious ritu-ximab-related side-effects. The authors concludedthat their data study confirms previous findings ofa favourable effect of rituximab in patients withCAS. However, they suggested that few patientswill obtain complete remission and, in most pa-tients, the effect will be transient.
Other reports also indicated that rituximab iseffective in achieving remissions in patients withidiopathic and secondary CAS.50–54
Plasma exchangeThere is some logic to the use of plasma exchangein those hemolytic anemias caused by IgM antibod-ies, since IgM has a predominantly intravasculardistribution, and experience has been favorablewith treatment of IgM-induced hyperviscosity syn-drome and hemostatic defects in Waldenstrom’smacroglobulinemia. However, the results inpatients with cold agglutinin syndrome have beenfar from impressive and improvement, when it oc-curred, has generally been only temporary. Fur-ther, dealing with cold agglutinins with a highthermal range necessitates special procedures toprevent autoagglutination during the procedure.
McLeod et al,55 reviewed 14 reported cases ofcold agglutinin syndrome that had been treatedwith plasma exchange. They stated that mostpatients responded with rapid and substantialreductions in antibody titer and that elevations ofhemoglobin level and/or decreases in transfusionrequirement were seen in many patients. However,the improvement in many cases was transient andapheresis therapy did not confer significant long-term benefit. Thus, the procedure appears to offera modest degree of temporary benefit in somepatients with cold agglutinin syndrome.
SplenectomySplenectomy is generally considered ineffective inpatients with cold agglutinin syndrome. McCurdyand Rath reported failure in one patient.56 Bellet al.57 described the outcome in three patients;two died of sepsis and one died of lymphoma. Pirof-sky15 described one patient who derived no clinicalbenefit.However, other investigatorshavedescribedmarked clinical improvement after splenectomy,particularly in patients whose antibody caused lysisof enzyme-treated red cells at 37 �C.33,58,59
Intravenous immunoglobulin (IVIG)Rather little data are available concerning the useof IVIG in cold agglutinin syndrome. A review ofcases of AIHA treated with IVIG was developed bythe University HealthSystem Consortium. Of the77 cases of AIHA evaluated, all but one were warmantibody AIHA.60 The exception was a patient withcold agglutinin syndrome who was refractory tothis therapy.61
Therapeutic failuresAlthough chronic idiopathic CAS is often a mild dis-order, there are a number of reports of more seri-ous disease and the failure of a variety oftherapeutic maneuvers. For example, Seldonet al62 reported a fatal case of acute low-titerwide-thermal-range cold agglutinin syndrome.
Cold antibody autoimmune hemolytic anemias 7
High-dose corticosteroids, cyclophosphamide, andplasmapheresis failed to control hemolysis. Simi-larly, Rousey and Smith63 reported a patient whohad idiopathic cold agglutinin syndrome with alow titer IgM cold agglultinin with anti-PR specific-ity, and who died because of refractory hemolysis.The patient required frequent transfusions (2–5units of RBC per week). All therapies that were at-tempted failed including prednisone, cyclophos-phamide, azathioprine, IVIG, plasmapheresis anddanazol. Serologic studies documented rising titersof the IgM cold agglutinin and immediately prior todeath, the thermal amplitude reached 37 �C.
Paroxysmal cold hemoglobinuria (PCH)
Incidence
Many reports emphasize that paroxysmal coldhemoglobinuria (PCH) is an unusual disease.Howard et al64 noted only two occurrences in 38years and 298,878 admissions at the Montreal Hos-pital. Becker65 noted that the diagnosis was madein only I patient in 20 years at the University of Chi-cago clinic, during which time 130,000 patientswere admitted to the hospital and 382,799 patientswere seen as out-patients. Pirofsky15 encounteredonly 2 occurrences over a I5-year period at Belle-vue Hospital in New York and at the University ofOregon hospitals and clinics. PCH was diagnosedin only 4 of 34 children with acute AIHA by Habibi8
and only one of 17 children reviewed by Buchananet al.66 Subsequent reports suggested a somewhathigher incidence. Dacie and Worlledge3 reported15 cases out of a total of 195 patients with AIHA(5.1 percent) and van Loghem4 and his associatesreported an incidence of 10 percent in their 168patients. Johnsen et al67 encountered 3 cases with-in a period of 7 months, and especially remarkableis the report of Bird et al68 who diagnosed this dis-order in 3 children within a period of just 16 days.
In more recent years, PCH has become recog-nized as being a relatively frequent cause of acutetransient AIHA in childhood. Sokol et al69 statedthat 17 of 42 (40%) of patients less than 16 yearsof age with AIHA had formed Donath-Landsteinerantibodies, and Gottsche et al70 reported an inci-dence in children with AIHA of 32% (22 of 68). Wynnet al71 reported six cases of children presentingwith PCH occurring in a 3-year period on thenorth-west of England, and Lambert et al72 re-ported six episodes of ‘‘biphasic hemolysis’’ in fourchildren over a 1-year period. Indeed, both Nordha-gen et al73 and Heddle74 have suggested that PCH isone of the most common causes of acute AIHA in
young children. The reason why acute transientPCH appears to be a more common type of child-hood AIHA than it was several decades ago is uncer-tain, but probably relates to greater awareness ofthe disorder and more frequent use of the Do-nath-Landsteiner test, especially in children withacute AIHA with hemoglobinuria.
Classification
Dacie75 has suggested that PCH should be classifiedinto three categories, i.e., chronic syphilitic,chronic non-syphilitic and acute transient. In mod-ern times, chronic PCH is very rare and almost allcases are of the acute and transient type. PCHmay also be classified as idiopathic or secondary.
Race, sex, and age
There does not appear to be any particular racialdistribution for the development of PCH. Puzzlinginformation is the fact that combined data in threereviews70,74,76 indicate that 52 of 77 children wereboys, a boy-to-girl ratio of 2.1:1.
Acute PCH typically presents in young children. Inthe series of Sokol et al77 the median age at presen-tation was 5 years. In one study, all of 22 patientswith PCH were less than 9 years old, no cases beingseen amongst 531 adults with AIHA.70 Although par-ticularly common in children, the disease affectspatients of all ages, and the patients in the seriesof Sokol et al77 ranged from 1 to 82 years.
Symptoms and signs
Characteristic clinical manifestations are indicatedin Table 4. The typical presentation is that of achild who, during the preceding 1–2 weeks, hadsuffered from an upper respiratory tract infec-tion.74,75,77,78 Usually the onset of hemolysis issignaled by a recurrence of fever and then the pas-sage of red-brown urine.75 Hemoglobinuria, jaun-dice and pallor were the most common clinicalfindings in 42 cases of acute PCH reported byHeddle.74 Of particular note is that hemoglobinuriawas found in 41 of the 42 cases in that report.Abdominal pain and fever were also common find-ings. Approximately 25 percent of cases presentedwith palpable liver and spleen. Hemoglobinuria fol-lowing exposure to cold is rare in acute PCH, andwas seen in only one of 52 patients reported bySokol et al.77 Nevertheless, hemoglobinuria maybe induced by exposure to cold.75,77
There are many descriptions in the older medicalliterature of Raynaud’s phenomenon occurring in
Symptoms and SignsPatient Usually a ChildHistory of a recent upper respiratory or ‘‘flu-like’’ illness.Acute onset of Illness
FeverMalaiseAbdominal Pain
Red or Red-Brown Urine is Frequently PresentJaundicePallor
Laboratory FindingsAnemia
Often SevereMay be Rapidly Progressive
Reticulocytosis (Reticulocytopenia in Some Patients)Abnormal RBC Morphology
Spherocytosis, anisocytosis, poikilocytosis, autoagglutination, Polychromatophilia.Erythrophagocytosis by neutrophils is commonly presentHemoglobinuria
Erythroid Hyperplasia or Normal Results in Bone MarrowLeukocytosis (Leukopenia in Some Patients)Platelet Count usually Normal or Elevated
8 L.D. Petz
association with attacks of hemoglobinuria in proba-ble cases of PCH.75 The fingers, toes, tip of the nose,lips, or ears may be blanched or become deeply cya-notic, and even gangrene has been described.
Hematologic findings
Typical laboratory findings in PCH are indicated inTable 4.
Hemoglobin
Severe and rapidly progressive anemia frequentlyoccurs. Levels of hemoglobin as low as 5 gm/dl oreven lower, have been quite commonlyreported.78–82 Sokol et al77 reported that the min-imum hemoglobin level had been less than 5 gm/dlin 17 of 51 cases (33%), and Heddle74 reviewed 42published cases and found that the admissionhemoglobin ranged from 2.5 to 12.5 gm/dl with amean value of 6.8 gm/dl. Gottsche et al 70 re-ported hemoglobin levels less than 5 gm/dl in sixout of 22 cases, the minimum being 4.4 gm/dl.
Reticulocytes
A relative reticulocytopenia occurs in some patientswith PCH at the time of presentation.71,72,77,78,80
Heddle74 reported that reticulocytopeniawaspromi-nent feature in four of the five cases seen in hermedical center.
Red blood cell morphology
The presence of spherocytes, anisocytosis, poikilo-cytosis, fragmented red cells, basophilic stippling,polychromatophilia, autoagglutination, erythro-phagocytosis, and nucleated erythrocytes have allbeen reported.15,57,67,77,78 Spherocytosis is particu-larly common in the peripheral blood of severelyaffected patients.67,68,75,77,78,80 Osmotic fragilityis increased commensurate with the spherocytosispresent,75 but has seldom been recorded.
Bone marrow examination
Bone marrow aspiration is not often performed,but, when carried out, generally reveals normal re-sults or erythroid hyperplasia.77
Leukocytes
Leukopenia followed by a neutrophil leukocytosiswith a shift to the left may occur during a paroxysminduced by cold, but usually normal white bloodcell counts or leukocytosis are reported duringacute attacks.
Platelets
Platelet counts seem to have been carried outinfrequently in cases of acute PCH.75 Where re-ported, the counts usually were normal, although
Cold antibody autoimmune hemolytic anemias 9
increased numbers have been reported in a fewpatients.
Diagnostic serologic tests for paroxysmalcold hemoglobinuria
The autoantibody associated with PCH is termed abiphasic hemolysin, that is to say, it sensitizes RBCsin the cold but only hemolyzes them when the RBCsreach 37 �C. The diagnostic test is the Donath Land-steiner (DL) test where normal RBCs are incubatedwith the patient’s serum at 0 �C (e.g., melting ice)and then moved to 37 �C for a further incubation.No lysis occurs following the incubation at 0 �C,and no lysis occurs if the incubation is carried outonly at 37 �C. The thermal amplitude of this anti-body is usually less than 20 �C, that is to say, theantibody will give a positive DL test only whenthe initial incubation is <20 �C; stronger results willoccur as the temperature of the initial incubation islowered. Rare patients have been described whenthe DL test is positive when the first incubationphase is as high as 32 �C, or their DL antibody wouldsensitize RBCs up to 37 �C, as detected by the IAT.
If positive results are obtained in the DL test,determination of the specificity of the autoanti-body is indicated. Almost all cases of PCH are asso-ciated with anti-P specificity, although rare reportsof other specificities said to be associated with PCHhave been reported. RBCs necessary for determin-ing anti-P specificity are rare but, with the assis-tance of reference laboratories, specificitytesting can be carried out. It should be noted thatthe DL test must be used to determine specificity.
A comparison of the typical characteristics ofthe antibody in CAS with the Donath-Landsteinerantibody of PCH is indicated in Table 5.
Erythrophagocytosis. Red cell adherence anderythrophagocytosis by neutrophils, rather thanmonocytes, is a prominent finding in acute tran-sient PCH,75,83 but is seen rarely in other forms ofimmune-mediated hemolytic anemia74 includingWAIHA.83 Heddle74 reported the phenomenon tobe present in 20% of all PCH cases and in 80% of
Table 5 Comparison of typical characteristics of the anLandsteiner antibody of paroxysmal cold hemoglobinuria.
Cold agg
Titer (4 �C) high (>50Thermal range high (>30Bithermic lysis (Donath-Landsteiner test) negativeImmunoglobulin class IgMSpecificity anti-I or
young children with acute transient PCH. Threereports devoted to this phenomenon are accompa-nied by photographs of erythrophagocytosis.84–86
Treatment and prognosis
The prognosis of acute PCH is excellent, althoughdeaths have been reported.76 Acute attacks arefrequently severe but the acute illness characteris-tically resolves spontaneously within a few days toseveral weeks after onset and rarely recurs.78
Accordingly, the patients require only supportivecare. During the acute phase of the disease, severeintravascular hemolysis is usual, and transfusion ofRBC may be necessary. Cortiocosteroids are oftengiven although their effectiveness is difficult toevaluate because of the transient nature of thehemolysis.
The Donath-Landsteiner antibody may persist inlow titer and with low thermal amplitude for longerperiod of time. Indeed, Dacie reported the persis-tence of the Donath-Landsteiner antibody for eightyears after a single episode of hemoglobinuria.58
More commonly, the titer and thermal range ofthe antibody fall rapidly, so that two to threemonths later it is no longer detectable.87
Combined cold and warm AIHA or‘‘Mixed’’ AIHA
In unusual instances, a patient will be found whohas serologic findings characteristic of warm anti-body AIHA while also having a cold agglutinin ofhigh titer and thermal amplitude, thus satisfyingthe criteria for both warm antibody AIHA and coldagglutinin syndrome.2 Such cases are frequently re-ferred to as ‘‘combined cold and warm AIHA’’ or‘‘mixed’’ AIHA, and reviews of AIHA generally in-clude such a category.88–93
Clinical characteristics
Sokol et al94 described 25 patients with mixedAIHA, and indicated that such patients accounted
tibody in cold agglutinin syndrome with the Donath-
lutinin Syndrome Donath-Landsteiner Antibody
0) moderate (<64)�C) moderate (< 20 �C)
positiveIgG
i anti-P
10 L.D. Petz
for about 7% of their cases of AIHA. The patientstended to have severe hemolysis which althoughusually responding well to initial treatment, ran achronic course with intermittent exacerbations,thus making overall management difficult.
Shulman et al95 reported that 12 of their 144patients (8.3%) with AIHA satisfied diagnostic cri-teria for both warm antibody AIHA and cold agglu-tinin syndrome. All patients had cold agglutininsthat reacted up to 37 �C, although in none ofthe patients was the cold agglutinin titer at 4 �Cgreater than 64. All 12 patients had severe hemo-lytic anemia that responded dramatically to corti-costeroid therapy, with the mean hemoglobinlevel increasing from 6.3 gm/dl to 12.9 gm/dl.However, similar to the patients reported by Sokolet al94 most of the patients reported by Shulmanet al95 were found to develop chronic hemolysis.Only in 2 patients was corticosteroid therapytapered off completely without a relapse of AIHA.Of the seven patients followed up who had chronichemolysis, five showed persistence of both IgGwarm autoantibodies and high-thermal amplitudecold autohemagglutinins, while two lost their coldautohemagglutinins and showed serologic findingsindistinguishable from those of warm antibodyAIHA.
Cautions concerning the diagnosis of coldand warm AIHA
The diagnosis of mixed cold and warm antibodyAIHA is sometimes made on the basis of inade-quate serologic studies.96,97 Appropriate charac-terization of the serum antibodies must bemade prior to determining the specific type ofAIHA present. An important source of potentialerror is suggested by our data2 which indicatethat 35% of patients with WAIHA have cold agglu-tinins reactive at 20 �C. However, a large major-ity of these are cold agglutinins are clinicallyinsignificant, and we found that only 5% reactedat 37 �C. Unless one can document a cold autoan-tibody with a high thermal amplitude (>30 �C) inassociation with a warm autoantibody, a diagnosisof cold and warm (‘‘mixed’’) AIHA is notwarranted.
Clinical course
Some authors suggest that patients with cold andwarm AIHA have a more severe onset and morechronic course than patients with other categoriesof AIHA. However, a comprehensive comparison
with the clinical course of patients with warm orcold AIHAs has not been made.
Secondary autoimmune hemolyticanemias
Autoimmune hemolytic anemias are classified assecondary for any of several reasons. One reasonis the association of AIHA with an underlying dis-ease with a frequency greater than can be ex-plained by chance alone. For example, allauthors agree that the incidence of warm anti-body AIHA is higher in patients with chronic lym-phocytic leukemia (CLL) and systemic lupuserythematosus (SLE) than in the generalpopulation.
Another criterion for categorizing a given case ofAIHA as secondary is the reversal of the hemolyticanemia simultaneously with the correction of theassociated disease. Ovarian tumors are a goodexample; well-documented cases of cure of theAIHA after surgical removal of the tumor have beenreported.98–101
Still another reason for suspecting a relationshipbetween the occurrence of AIHA and an associateddisease consists of evidence of immunologic aber-ration as part of the underlying disorder, especiallyif the associated disease is thought to have anautoimmune pathogenesis. However, when thecoexistence of two immunologic disorders in a sin-gle patient is infrequent, the pathogenetic signifi-cance is conjectural and interpretations vary. Forexample, rheumatoid arthritis is an extremelycommon disorder and reports of an associated AIHAare infrequent, so the combination may well becoincidental.99 Most cases of secondary CAS areassociated with infectious agents or hematologicmalignancies.
CAS and infectious diseases
Mycoplasma pneumonia infection and infectiousmononucleosis are the most common infectiousdiseases associated with CAS. Less commonly, coldagglutinin syndrome may be associated with rubel-la, varicella, cytomegalovirus, Legionnaires’ Dis-ease and HIV infection.1
Lymphoproliferative disorders
The records of 637 patients with lymphoprolifera-tive disorders and 346 patients with myeloprolifera-tive disorders were retrospectively analyzed for
Cold antibody autoimmune hemolytic anemias 11
the presence of co-existent autoimmune derange-ments.102 The frequency of autoimmune perturba-tions in lymphoproliferative diseases (51 cases;8.0%) was significantly higher than in myeloprolif-erative diseases (six cases; 1.7%; p < 0.0001). Rheu-matic disorders, autoallergic hematologicalmanifestations and other organ-specific autoim-mune derangements were responsible for aboutone third each of the observed disturbances. Thesedata are similar to those of Miller103 who observeddiffuse connective tissue diseases and autoimmunehemolytic anemias in 6.8% of his lymphomapatients.
AIHA is the most common autoimmune diseaseassociated with CLL and CLL is the most commonof the known causes of AIHA.104,105 Various reportsindicate that AIHA occurs in 5–37% of CLLpatients.105–111 The figures are highly dependenton the duration and extent of disease.105
In the great majority of cases of CLL compli-cated by AIHA the responsible autoantibodies havebeen of the warm type but there have been somereports of cold antibody AIHA in association withCLL.112–114 Ruzickova et al115 reported a patientwith a 7-year history of idiopathic CAS in whomB-CLL subsequently developed. They determinedthat the B-CLL had developed from the patient’scold agglutinin-producing B-cell population.
AIHA is an uncommon occurrence in patientswith Hodgkin’s disease. In 1967 Eisner et al116 re-ported the incidence of AIHA in a series of 219 pa-tients was 2.7%. With one exception, all of thepatients had advanced disease at the time ofdetection of the positive DAT. Levine et al117 re-viewed 71 cases of Hodgkin’s disease and foundthree who had AIHA (4.2%).
Only a minority of cases of secondary AIHA innon-Hodgkin’s lymphoma are associated with coldantibodies although a number of reports haveemphasized this association.118–124 Economopouloset al118 reviewed 370 patients with NHL and foundthat 23 (6.2%) had AIHA, 4 of whom (1.1%) had coldreacting antibodies.
The relationship of CAS and Waldenstrom’smacroglobulinemia
A number of authors have noted similarities oflaboratory findings in patients with CAS andpatients who have monoclonal IgM proteins with-out cold agglutinin activity. In both instances,bone marrow findings may consist of increasednumbers of abnormal lymphoid and plasmacells.125 Schubothe12 performed serial examina-tions of the bone marrow with quantitative differ-
ential counts in 14 patients with CAS. The numberof lymphoid cells was between 10 and 24 percentin 5 cases, between 26 and 33 percent in 3 casesand between 41 and 95 percent in 6 cases. In sev-eral patients, the lymphoid cells increased duringthe course of the disease. On the basis of suchfindings some authors feel that chronic CAS isprobably a variant of Waldenstrom’s macroglobu-linemia126 in which the IgM M-component has coldagglutinin activity.12,127,128
Non-hematologic malignancies and coldagglutinin syndrome
The association of nonhematologic malignant dis-orders with CAS is uncommon.88,129–131 Wortmanet al129 reported seven patients who had a varietyof nonhematologic malignancies including squa-mous cell carcinoma of the lung, metastaticadenocarcinoma of the adrenal, metastatic adeno-carcinoma of the colon, a mixed parotid tumor, acarcinoid tumor of the ampulla of Vater, carci-noma of the larynx, and a thoracic neurinoma.The high frequency of carcinomas in patients inthe age group in which cold agglutinin syndromeoccurs suggests that the relationship may be fortu-itous in at least some instances. That this is nottrue in all cases is indicated by the patient re-ported by Lands and Foust130 in whom resolutionof the AIHA occurred after excision of a renal cellcarcinoma.
Secondary paroxysmal cold hemoglobinuria
In the 19th and early 20th centuries PCH was fre-quently found in association with latent syphilis,but with the development of effective antibiotictherapy for syphilis, this association has becomerare,75,77 although not unknown.77,132–138 Recentreports generally describe an acute transienthemolytic anemia often associated with a upperrespiratory infection or a ‘‘flu-like illness’’ ofundefined etiology. However, PCH has been asso-ciated with a very wide range of specific infec-tious agents including measles, mumps, chickenpox, Coxsackie A9, parvovirus, cytomegalovirus,infectious mononucleosis, Haemophilus influen-zae, Mycoplasma pneumoniae, Klebsiella pneu-monia, adenovirus and influenza virus A.1 Also,PCH has rarely been associated with other disor-ders, including non-Hodgkin’s lymphoma139,140
and oat cell carcinoma.141 Some patients haveno history of an antecedent illness, and they ap-pear healthy until the onset of the acuteparoxysm.68
12 L.D. Petz
Practice points
� Immune hemolytic anemias are divided intodistinctive categories which have differingclinical manifestations, prognosis andtherapy.� Cold antibody autoimmune hemolytic ane-mias are comprised of cold agglutinin syn-drome (CAS) and paroxysmal coldhemoglobinuria (PCH).� CAS occurs in middle-aged or elderly patientsand is caused by an IgM antibody that is maxi-mally reactive in the cold but which reacts toat least 30 �C in vitro.� PCH occurs primarily in children and iscaused by an IgG biphasic hemolysin that isdemonstrated by incubation in the cold fol-lowed by incubation at 37 �C in the presenceof complement.� Both CAS and PCH may be primary (idio-pathic) or secondary to other disorders.� Therapy of CAS primarily consists of avoid-ance of cold, immunosuppressive drugs andrituximab.� Therapy of PCH consists of supportive caresince the disease is generally self-limiting,albeit after a frequently stormy course.
References
1. Petz LD, Garratty G. Classification and Clinical Character-istics of Autoimmune Hemolytic Anemias. In: Petz LD,Garratty G, editors. Immune Hemolytic Anemias. seconded. Philadelphia: Churchill Livingstone; 2004. p. 61–131.
2. Petz LD, Garratty G. Acquired Immune Hemolytic Anemias.first ed. New York: Churchill; 1980.
3. Dacie JV, Worlledge SM. Auto-immune hemolytic anemias.In: Brown EB, Moore CV, editors. Progr.Hematol. sixthed. New York: Grune and Stratton; 1969. p. 82–120.
4. Van Loghem JJ, Van der Hart M, Dorfmeier H. Serologicalstudies in acquired hemolytic anemia. Proceedings of theSixth International Congress of the International Societyof Hematology. New York: spring; 1958.
5. Dausset J, Colombani J. The serology and the prognosis of128 cases of autoimmune hemolytic anemia. Blood1959;14:1280.
6. Worlledge S. Auto-immunity and blood diseases. Practi-tioner 1967;199:171–9.
7. Dacie JV. Auto-immune haemolytic anaemia (AIHA): cold-antibody syndromes. I: idiopathic types: clinical presen-tation and haematological and serological findings. TheHaemolytic Anaemias. third ed. The Auto-Immune Hae-molytic Anaemias. New York: Churchill Livingstone; 1992.p. 210–39.
8. Habibi B, Homberg JC, Schaison G, Salmon C. Autoimmunehemolytic anemia in children. A review of 80 cases. Am JMed 1974;56:61–9.
9. Zupanska B, Lawkowicz W, Gorska B, et al. Autoimmunehaemolytic anaemia in children. Br J Haematol1976;34:511–20.
11. Nydegger UE, Kazatchkine MD, Miescher PA. Immunopath-ologic and clinical features of hemolytic anemia due tocold agglutinins. Seminars Hematol 1991;28:66–77.
12. Schubothe H. The cold hemagglutinin disease. SeminHematol 1966;3:27–47.
13. Crisp D, Pruzanski W. B-cell neoplasms with homogeneouscold-reacting antibodies (cold agglutinins). Am J Med1982;72:915–22.
14. Ferriman DG, Dacie JV, Keele KD, Fullerton JM. Theassociation of Raynaud’s phenomena, chronic haemolyticanaemia, and the formation of cold antibodies. Quart JMed 1951;20:275.
15. Pirofsky B. Autoimmunization and the Autoim-mune Hemolytic Anemias. Baltimore: Williams & Wilkins;1969.
16. Lauchli S, Widmer L, Lautenschlager S. Cold agglutinindisease–the importance of cutaneous signs. Dermatology2001;202:356–8.
17. Mitchell AB, Pergrum GD, Gill AM. Cold agglutinin diseasewith Raynaud’s phenomenon. Proc R Soc Med 1974;67:113–5.
18. Poldre P, Pruzanski W, Chiu HM, Dotten DA. Fulminantgangrene in transient cold agglutinemia associated withEscherichia coli infection. Can Med Assoc J 1985;132:261–3.
19. Stone MS, Piette WW, Davey WP. Cutaneous necrosis atsites of transfusion: cold agglutinin disease. J Am AcadDermatol 1988;19(2 Pt 1):356–7.
20. Zouboulis CC, Gollnick H, Orfanos CE. Essential thrombocy-themia and idiopathic cold agglutinin disease: two addi-tional disorders associated with intravascular coagulationnecrosis of the skin. J Am Acad Dermatol 1992;27(2 Pt 1):277.
21. Barth JH. Infectious mononucleosis (glandular fever) com-plicated by cold agglutinins, cold urticaria and leg ulcer-ation. Acta Derm Venereol 1981;61:451–2.
23. DeLange JA, Eernisse GJ, Veltkamp JJ. Cold agglutininsand the Coulter Counter Model S. Am J Clin Pathol1972;58:599–600.
24. Lawrence C, Zozicky O. Spurious red-cell values with theCoulter Counter. N Engl J Med 1983;309:925–6.
25. Petrucci JV, Dunne PA, Chapman CC. Spurious erythrocyteindices as measured by the model S Coulter counter due tocold agglutinins. Am J Clin Pathol 1971;56:500–2.
26. Hattersley PG, Gerard PW, Caggiano V, Nash DR.Erroneous values on the Model S Coulter Counter due tohigh titer cold autoagglutinins. Am J Clin Pathol 1971;55:442–6.
27. Roelcke D. Cold agglutination. Transfus Med Rev1989;3:140–66.
28. Leddy JP, Swisher SN. Acquired Immune Hemolytic Disor-ders (Including Drug-Induced Immune Hemolytic Anemia).In: Samter M, editor. Immunologic Diseases. 3 ed. Bos-ton: Little Brown & Co.; 1978. p. 1187.
29. Schreiber AD, Herskovitz BS, Goldwein M. Low-titer cold-hemagglutinin disease. Mechanism of hemolysis andresponse to corticosteroids. N Engl J Med 1977;296:1490–4.
Cold antibody autoimmune hemolytic anemias 13
30. Rosse WF. Correlation of in vivo and in vitro measurementsof hemolysis in hemolytic anemia due to immune reac-tions. Prog Hematol 1973;8:51–75.
31. Fudenberg HH, Kunkel HG. Physical properties of the redcell agglutinins in acquired hemolytic anaemia. J Exp Med1957;1:689.
32. Schwartz T, Jager BV. Cryoglobulinemia and Raynaud’ssyndrome in a case of chronic lymphocytic leukemia.Cancer 1949;2:319.
33. Dacie JV. Auto-immune haemolytic anaemias (AIHA):treatment. third ed. The Haemolytic Anaemias; TheAuto-Immune Haemolytic Anaemias. New York: ChurchillLivingstone; 1992. p. 452–520.
34. Pisciotta AV. Cold hemagglutination in acute and chronichemolytic syndromes. Blood 1955;10:295.
35. Firkin BG, Blackwell JB, Johnston GAW. Essential cryo-globulinaemia and acquired haemolytic anaemia due tocold agglutinins. Australian Ann Med 1959;8:151.
36. Nanan R, Scheurlen W, Gerlich M, Huppertz HI. Severe low-titer cold-hemagglutinin disease responsive to steroidpulse therapy. Ann Hematol 1995;71:101–2.
37. Lahav M, Rosenberg I, Wysenbeek AJ. Steroid-responsiveidiopathic cold agglutinin disease: a case report. ActaHaematol 1989;81:166–8.
38. WorlledgeS. ImmuneHaemolyticAnaemias. In:HardistyRM,Weatherall DJ, editors. Blood and its Disor-ders. Oxford: Blackwell; 1982. p. 479–513.
39. Worlledge SM, Brain MC, Cooper AC, Hobbs JR, Dacie JV.Immmunosuppressive drugs in the treatment of autoim-mune haemolytic anaemia. Proc R Soc Med 1968;61:1312–5.
40. Hippe E, Jensen KB, Olesen H, Lind K, Thomsen PE.Chlorambucil treatment of patients with cold agglutininsyndrome. Blood 1970;35:68–72.
41. Jacobs A. Cold agglutinin hemolysis responding to fludar-abine therapy. Am J Hematol 1996;53:279–80.
42. Petz LD, Garratty G. Drug-Induced Immune HemolyticAnemia. In: Petz LD, Garratty G, editors. Immune Hemo-lytic Anemias. second ed. Philadelphia: Churchill Living-stone; 2004. p. 261–317.
43. Pescovitz MD. Rituximab, and anti-CD20 monoclonal anti-body: history and mechanism of action. American Journalof Transplantation 2006;6:859–66.
44. Lee E, Zamkoff KW, Gentile TC, Zimrin A. Rituxan in thetreatment of auto-immune hemolytic anemia (AIHA).Blood 2000;96:596a–7a.
46. Berentsen S, Tjonnfjord GE, Brudevold R, et al. Favour-able response to therapy with the anti-CD20 monoclonalantibody rituximab in primary chronic cold agglutinindisease. Br J Haematol 2001;115:79–83.
47. Pulik M, Genet P, Lionnet F, Touahri T. Treatment ofprimary chronic cold agglutinin disease with rituximab:maintenance therapy may improve the results. Br JHaematol 2002;117:998–9.
48. Berentsen S, Ulvesstad E, Gjertsen BT, et al. Rituximab forprimary chronic cold agglutinin disease: a prospectivestudy of 37 courses of therapy in 27 patients. Blood2004;103:2925–8.
49. Schollkopf C, Kjeldsen L, Bjerrum OW, et al. Rituximab inchronic cold agglutinin disease: a prospective study of 20patients. Leukemia & Lymphoma 2006;47:253–60.
50. Layios N, Van Den NE, Jost E, Deneys V, Scheiff JM, FerrantA. Remission of severe cold agglutinin disease afterRituximab therapy. Leukemia 2001;15:187–8.
51. Mori A, Tamaru J, Sumi H, Kondo H. Beneficial effectsof rituximab on primary cold agglutinin disease refrac-tory to conventional therapy. Eur J Haematol 2002;68:243–6.
52. Engelhardt M, Jakob A, Ruter B, Trepel M, Hirsch F,Lubbert M. Severe cold hemagglutinin disease (CHD)successfully treated with rituximab. Blood 2002;100:1922–3.
53. Cohen Y, Polliack A, Zelig O, Goldfarb A. Monotherapy withrituximab induces rapid remission of recurrent cold agglu-tinin-mediated hemolytic anemia in a patient with indolentlymphoplasmacytic lymphoma. Leukemia & Lymphoma2007;42:1405–8.
54. Kotani T, Takeuchi T, Kawasaki Y, et al. Successfultreatment of cold agglutinin disese with anti-CD20 anti-body (rituximab) in a patient with systemic lupus erythe-matosus. Lupus 2007;15:683–5.
55. McLeod BC, Strauss RG, Ciavarella D, et al. Management ofhematological disorders and cancer. J Clin Apheresis1993;8:211–30.
56. McCurdy PR, Rath CE. Splenectomy in hemolytic anemia:results predicted by body scanning after injection of Cr51-tagged red cells. New Engl J Med 1958;259:459.
57. Bell CA, Zwicker H, Sacks HJ. Autoimmune hemolyticanemia: routine serologic evaluation in a general hospitalpopulation. Am J Clin Pathol 1973;60:903–11.
58. Dacie JV. The Haemolytic Anaemias. second ed. Part II.Auto-Immune Haemolytic Anaemias. London: J. & A Chur-chill Ltd.; 1962.
59. Evans RS, Bingham M, Turner E. Autoimmune hemolyticdisease: Observations of serologic reactions and diseaseactivity. Ann NY Acad Sci 1965;124:422.
61. Besa EC. Rapid transient reversal of anemia and long-termeffects of maintenance intravenous immunoglobulin forautoimmune hemolytic anemia in patients with lympho-proliferative disorders. Am J Med 1988;84:691–8.
62. Seldon M, Isbister JP, Raik E, Biggs JC. A fatal case of coldautoimmune hemolytic anemia. Am J Clin Pathol1980;73:716–7.
63. Rousey SR, Smith RE. A fatal case of low titer anti-PR coldagglutinin disease. Am J Hematol 1990;35:286–7.
66. Buchanan GR, Boxer LA, Nathan DG. The acute andtransient nature of idiopathic immune hemolytic anemiain childhood. J Pediatr 1976;88:780–3.
67. Johnsen HE, Brostrphom K, Madsen M. Paroxysmal coldhaemoglobinuria in children: 3 cases encountered within aperiod of 7 months. Scand J Haematol 1978;20:413–6.
68. Bird GW, Wingham J, Martin AJ, et al. Idiopathic non-syphilitic paroxysmal cold haemoglobinuria in children. JClin Pathol 1976;29:215–8.
69. Sokol RJ, Hewitt S, Stamps BK, Hitchen PA. Autoimmunehaemolysis in childhood and adolescence. Acta Haematol1984;72:245–57.
70. Gottsche B, Salama A, Mueller-Eckhardt C. Donath-Land-steiner autoimmune hemolytic anemia in children. A studyof 22 cases. Vox Sang 1990;58:281–6.
71. Wynn RF, Stevens RF, Bolton-Maggs PH, Schwe K, Will AM.Paroxysmal cold haemoglobinuria of childhood: a review ofthe management and unusual presenting features of sixcases. Clin Lab Haematol 1998;20:373–5.
14 L.D. Petz
72. Lambert T, Noblins M, Tchernia G. Biphasic hemolysinhemolytic anemia in children. Am J Hematol 1993;43:77.
73. Nordhagen R, Stensvold K, Winsnes A, Skyberg D, Storen A.Paroxysmal cold haemoglobinuria. The most frequentacute autoimmune haemolytic anaemia in children? ActaPaediatr Scand 1984;73:258–62.
77. Sokol RJ, Booker DJ, Stamps R. Erythropoiesis: Paroxysmalcold haemoglobinuria: A clinico-pathological study ofpatients with a positive Donath-Landsteiner test. Hematol1999;4:137–64.
78. Wolach B, Heddle N, Barr RD, Zipursky A, Pai KR,Blajchman MA. Transient Donath-Landsteiner haemolyticanaemia. Br J Haematol 1981;48:425–34.
85. Gelfand EW, Abramson N, Segel GB, Nathan DG. Buffy-coatobservations and red-cell antibodies in acquired hemolyticanemia. N Engl J Med 1971;284:1250–2.
86. Hernandez JA, Steane SM. Erythrophagocytosis by seg-mented neutrophils in paroxysmal cold hemoglobinuria.Am J Clin Pathol 1984;81:787–9.
87. Worlledge S. Immune Haemolytic Anemias. In: HardistyDJ, Weatherall DJ, editors. Blood and Its Disor-ders. Oxford: Blackwell; 1974. p. 714.
88. Sokol RJ, Booker DJ, Stamps R. Erythrocyte autoantibod-ies, autoimmune haemolysis, and carcinoma. J Clin Pathol1994;47:340–3.
89. Sokol RJ, Hewitt S. Autoimmune hemolysis: a criticalreview. Crit Rev Oncol Hematol 1985;4:125–54.
91. Hsu HC, Lin CK, Chau WK, Hu HY, Liu SM. Combined cold-and warm-antibody autoimmune hemolytic anemia–review of the literature and a case report. Gaoxiong YiXue Ke Xue Za Zhi 1989;5:350–6.
95. Shulman IA, Branch DR, Nelson JM, Thompson JC, Saxena S,Petz LD. Autoimmune hemolytic anemia with both cold andwarm autoantibodies. JAMA 1985;253:1746–8.
96. Koduri PR, Singa P, Nikolinakos P. Autoimmune hemolyticanemia in patients infected with human immunodeficiencyvirus-1. Am J Hematol 2002;70:174–6.
97. Nusbaum NJ, Khosla S. Autoimmune hemolytic anemia withboth cold and warm autoantibodies. JAMA 1985;254:1175–6.
98. Barry KG, Crosby WH. Auto-immune hemolytic anemiaarrested by removal of an ovarian teratoma: review of theliterature and report of a case. Ann Intern Med 1957;47:1002–7.
99. Dawson MA, Talbert W, Yarbro JW. Hemolytic anemiaassociated with an ovarian tumor. Am J Med 1971;50:7552–6.
100. de Bruyere M, Sokal G, Devoitille JM, Fauchet-Dutrieux MC,De SV. Autoimmune haemolytic anaemia and ovariantumour. Br J Haematol 1971;20:83–94.
101. Payne D, Muss HB, Homesley HD, Jobson VW, Baird FG.Autoimmune hemolytic anemia and ovarian dermoid cysts:case report and review of the literature. Cancer1981;48:721–4.
102. Duhrsen U, Augener W, Zwingers T, Brittinger G. Spectrumand frequency of autoimmune derangements in lympho-proliferative disorders: analysis of 637 cases and compar-ison with myeloproliferative diseases. Br J Haematol1987;67:235–9.
103. Miller DG. The association of immune disease and malig-nant lymphoma. Ann Intern Med 1967;66:507–21.
105. Diehl LF, Ketchum LH. Autoimmune disease and chroniclymphocytic leukemia: autoimmune hemolytic anemia,pure red cell aplasia, and autoimmune thrombocytopenia.Semin Oncol 1998;25:80–97.
106. Mauro FR, Foa R, Cerretti R, et al. Autoimmune hemolyticanemia in chronic lymphocytic leukemia: clinical, thera-peutic, and prognostic features. Blood 2000;95:2786–92.
107. Hansen MM. Chronic lymphocytic leukaemia. Clinical stud-ies based on 189 cases followed for a long time. Scand JHaematol Suppl 1973;18:3–286.
108. Bartal A, Bentwich Z, Manny N, Izak G. Ethnical and clinicalaspects of chronic lymphocytic leukemia in Israel: a surveyon 288 patients. Acta Haematol 1978;60:161–71.
109. Seaman AJ, Koler RD, Pirofsky B, et al. Incidence andtreatment of acquired hemolytic anemia complicatingchronic lymphocytic leukemia. Proceedings of the SixthInternational Congress of the International Society ofHematology 1956:857.
110. Robertson TI. Complications and causes of death in B cellchronic lymphocytic leukaemia: a long term study of 105patients. Aust N Z J Med 1990;20:44–50.
111. De Rossi G, Granati L, Girelli G, et al. Incidence andprognostic significance of autoantibodies against erythro-cytes and platelets in chronic lymphocytic leukemia (CLL).Nouv Rev Fr Hematol 1988;30:403–6.
112. Greally JF, Whelan CA, O’Connell L, O’Gorman D. Coldagglutinins in a case of chronic lymphatic leukaemia. Astudy of the lymphocyte surface. Acta Haematol1977;57:206–10.
113. Schwartz TB, Jager BV. Cryoglobulinemia and Raynaud’ssyndrome in a case of chronic lymphocytic leukemia.Cancer 1949;2:319–28.
Cold antibody autoimmune hemolytic anemias 15
114. Dacie JV. The Haemolytic Anaemias. third ed. Secondaryor Symptomatic Haemolytic Anaemias. New York: Chur-chill Livingstone; 1995.
115. Ruzickova S, Pruss A, Odendahl M, et al. Chronic lympho-cytic leukemia preceded by cold agglutinin disease:intraclonal immunoglobulin light-chain diversity in V(H)4-34 expressing single leukemic B cells. Blood 2002;100(9):3419–22.
116. Eisner E, Ley AB, Mayer K. Coombs’-positive hemolyticanemia in Hodgkin’s disease. Ann Intern Med 1967;66:258–73.
117. Levine AM, Thornton P, Forman SJ, et al. Positive Coombstest in Hodgkin’s disease: significance and implications.Blood 1980;55:607–11.
118. Economopoulos T, Stathakis N, Constantinidou M, Papa-georgiou E, Anastassiou C, Raptis S. Cold agglutinin diseasein non-Hodgkin’s lymphoma. Eur J Haematol 1995;55:69–71.
120. Costea N, Yakulis VJ, Heller P. The mechanisms ofinduction of cold agglutinins by M pneumoniae. Blood1968;34:829. Ref Type: Abstract.
121. Bassan R, Pronesti M, Buzzetti M, et al. Autoimmunity andB-cell dysfunction in chronic proliferative disorders oflarge granular lymphocytes/natural killer cells. Cancer1989;63:90–5.
122. Liaw YS, Yang PC, Su IJ, Kuo SH, Wang CH, Luh KT. Mucosa-associated lymphoid tissue lymphoma of the lung withcold-reacting autoantibody-mediated hemolytic anemia.CHEST 1994;105:288–90.
123. Pascali E. Monoclonal gammopathy and cold agglutinindisease in non-Hodgkin’s lymphoma. Eur J Haematol1996;56:114–5.
124. Bar BM, Reijnders FJ, Keuning JJ, Bal H, van Beek M.Primary malignant lymphoma of the uterine cervix associ-ated with cold-reacting autoantibody-mediated hemolyticanemia. Acta Haematol 1986;75:232–5.
125. Ritzmann SE, Levin WC. Cold agglutinin disease: A type ofprimary macroglobulinemia: a new concept. Tex Rep BiolMed 1962;20:236.
126. Johnson SA, Oscier DG, Leblond V. Waldenstrom’s macro-globulinaemia. Blood Reviews 2002;16:175–84.
127. Pruzanski W, Shumak KH. Biologic activity of cold-reactingautoantibodies (first of two parts). N Engl J Med1977;297:538–42.
128. Pruzanski W, Shumak KH. Biologic activity of cold-reactingautoantibodies (second of two parts). N Engl J Med1977;297:583–9.
129. Wortman J, Rosse W, Logue G. Cold agglutinin autoimmunehemolytic anemia in nonhematologic malignancies. Am JHematol 1979;6:275–83.
130. Lands R, Foust J. Renal cell carcinoma and autoimmunehemolytic anemia. South Med J 1996;89:444–5.
131. Cao L, Kaiser P, Gustin D, Hoffman R, Feldman L. Coldagglutinin disease in a patient with uterine sarcoma. Am JMed Sci 2000;320:352–4.
133. Patel M, Durao H, Govender Y. Paroxysmal cold haemo-globinuria coexisting with cold agglutinins in a patient withsyphilis resulting in peripheral gangrene: a case report.East Afr Med J 1993;70:526–7.
134. Carswell J, Eloff P, Edkins P, Willmore A. Paroxysmal coldhaemoglobinuria. A case report. S Afr Med J 1979;55:891–2.
135. Nelson MG, Nicholl B. Paroxysmal cold haemoglobinuria.Irish J Med Sci 1960;6th series:49–57.
136. Wood JK. Paroxysmal cold haemoglobinuria in Jamaica. Acase report. West Indian Med J 1968;17:175–9.
137. Shah AA, Desai AB. Paroxysmal cold hemoglobinuria (casereport). Indian Pediatr 1977;14:219–21.
138. Kumar ND, Sethi S, Pandhi RK. Paroxysmal cold haemoglo-binuria in syphilis patients. Genitourin Med 1993;69:76.
139. Sharara AI, Hillsley RE, Wax TD, Rosse WF. Paroxysmal coldhemoglobinuria associated with non-Hodgkin’s lymphoma.South Med J 1994;87:397–9.
141. Lippman SM, Winn L, Grumet FC, Levitt LJ. Evans’syndrome as a presenting manifestation of atypical parox-ysmal cold hemoglobinuria. Am J Med 1987;82:1065–72.
