Alveolar Hemorrhage: An Underdiagnosed Complication of Treatment with Glycoprotein IIb/IIIa Inhibitors

Please download to get full document.

View again

of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
Objective: Alveolar hemorrhage (AH) is a rare complication of treatment with GP IIb/IIIa inhibitors. Hemoptysis, a constant sign, lacks in specificity, and may occur in confounding syndromes such as pulmonary edema, pulmonary infarction, and
Document Share
Document Tags
Document Transcript
  C  2006, the AuthorsJournal compilation  C  2006, Blackwell Publishing, Inc. Alveolar Hemorrhage:  An Underdiagnosed Complication of Treatment withGlycoprotein IIb/IIIa Inhibitors SAID B. ISKANDAR, M.D., 1 EHAB S. KASASBEH, M.D., 2 BASSAM K. MECHLEB, M.D., 1 ISRAEL GARCIA, M.D., F.A.C.C., 1 ANN JACKSON, M.D., F.A.C.C., 1 STEPHEN FAHRIG, M.D., F.A.C.C., 1 KAISS ALBALBISSI, M.D., 2 and PHILLIP D. HENRY, M.D., F.A.C.C. 1 From the  1  Division of Cardiology;  2  Department of Internal Medicine, Quillen College of Medicine, East Tennessee State University, JohnsonCity, Tennessee Objective:  Alveolarhemorrhage(AH)isararecomplicationoftreatmentwithGPIIb/IIIainhibitors.Hemoptysis,aconstant sign, lacks in specificity, and may occur in confounding syndromes such as pulmonary edema, pulmonaryinfarction, and pneumonia. Nonspecific symptoms and signs often delay the diagnosis, thereby allowing seriousor even fatal disease progression. Here, we performed a large-scale retrospective analysis to define the incidenceand risk factors of AH in the setting of GP IIb/IIIa inhibitors therapy.  Background:  Randomized controlled trials demonstrate that treatment with glycoprotein IIb/IIIa (GP IIb/IIIa)inhibitors may improve the outcome of acute myocardial infarction (AMI) and angioplastic procedures. However,thistreatmentmayrarelyleadtoseverehemorrhagiccomplications,inparticularAH.Unfortunately,theincidenceand risk factors of AH remain poorly defined.  Methods:  We reviewed for the period extending from August 1998 to January 2005 consecutive histories of AMI  patients receiving coronary arteriography and treatment with either eptifibatide or abciximab. Concomitantlyadmitted AMI patients not treated with GP IIb/IIIa inhibitors were reviewed and served as a control group.The diagnosis of AH required the demonstration of typical symptoms and signs including dyspnea, hemoptysis,arterial hypoxemia, pulmonary radiographic changes, and, when available, bronchoscopic signs for AH. Potentialcovariates including pulmonary disease, pulmonary hypertension, smoking, and use of other anticoagulant or antiplatelet agents were evaluated.  Results: Sixof1,810patients(0.33%)receivingeptifibatideandfiveof3,648patients(0.14%)receivingabciximabexhibited typical symptoms and signs of AH. Contrarily, only one of 4,136 patients (0.025%) receiving no GP IIb/IIIa inhibitors presented with similar symptoms and signs. There was no fatal outcome, though two patientsrequired blood transfusions. Statistically significant differences were found between control patients and patientsreceivingeptifibatidealone(P = 0.004).Therewasalsoasignificantdifferencebetweenuntreatedpatientsandthosereceiving eptifibatide and abciximab (P = 0.017). No differences were found between eptifibatide and abciximab-treated patients (P = 0.19) or between abciximab and untreated control patients (P = 0.105). Conclusions:  AH is a rare complication of treatment with GP IIb/IIIa inhibitors. Its incidence ranged from 0.14%in patients treated with abciximab to 0.33% in those receiving eptifibatide. Compared to a control group, patientstreated with GP IIb/IIIa inhibitors had a statistically increased risk for AH.  (J Interven Cardiol 2006;19:356–363) Introduction The pivotal role played by platelet activation andaggregation in the development of acute myocardial Address for reprints: Said B. Iskandar, M.D., Division of Cardiol-ogy, Quillen College of Medicine, East Tennessee State University,2 Professional Park Drive, Suite 15, Johnson City, TN 37604. Fax:423-722-0002; e-mail: infarction (AMI) and in the outcome of percutaneouscoronary interventions has promoted the developmentof several inhibitors targeting the platelet glycoprotein(GP) IIb/IIIa receptor. 1 Abciximab (ReoPro; Eli Lilly;Indianapolis, IN), a chimeric human-murine mono-clonal antibody binding to the platelet glycoproteinIIb/IIIa receptor was the first agent to undergo clinicalevaluation. 2 356 Journal of Interventional Cardiology Vol. 19, No. 4, 2006  ALVEOLAR HEMORRHAGE AND GP IIB/IIIA INHIBITORS Subsequently, nonimmunological agents includingtirofiban,anonpeptide(Aggrastat,Merck  , WhitehouseStation, NJ) and eptifibatide, a cyclic heptapeptide(Integrilin, Millennium Pharmaceuticals, Inc., Cam-bridge, MA) were developed and evaluated in random-ized controlled trials. These trials demonstrated thattreatment with GP IIb/IIIa inhibitors improve the out-comeofpatientsundergoingcoronaryangioplastywithor without stent placement 3–8 or presenting with non-ST segment elevation myocardial infarction and unsta-ble angina. 7 , 9 ThemostcommonadversereactionwithGPIIb/IIIainhibitors therapy is bleeding. Major and minor bleed-ing has been reported to occur in 4.4–10.8% and 10.5–14.2% and the requirement for blood transfusion wasreported in 5.5–12.8%. Death attributable to bleedingwas described in 0.9%. 10 Major bleeding was definedas intracranial hemorrhage or bleeding determining ahemoglobin drop of at least 5 g/dL. Bleeding mostfrequently arose from arterial access sites for cardiaccatheterization, the gastrointestinal tract or the geni-tourinary tract. Alveolar hemorrhage (AH), comparedwith other hemorrhagic syndromes, is difficult to diag-nose and may be easily overlooked. In one study, theunivariate predictors of bleeding complications wereold age, low body weight, elevated baseline creatinine,elevated activated partial thromboplastin time (aPTT),history of diabetes mellitus, congestive heart failure,peripheral vascular disease (PVD), and AMI. By mul-tivariate analysis, only age, history of PVD, and pro-longed aPTT were retained as predictors. 10 We undertook a review of the records of all patientswhoreceivedabciximaboreptifibatideduringcoronaryinterventionproceduresatourinstitutionsbetweenAu-gust 1998 and January 15, 2005. We also reviewedfor the same period the records of consecutive patientswho underwent coronary interventions or were admit-ted with AMI, but did not receive abciximab or eptifi-batide.Thesepatientsservedasuntreatedcontrols.Thepurposeofthisstudywastodeterminetheincidenceof AH in patients who received eptifibatide or abciximaband to compare them to an appropriate control group. Methods The medical records of 5,458 consecutive patientswhowereadmittedwiththediagnosisofAMIandweretreated with either eptifibatide or abciximab were re-viewed.Therecordsoftwoinstitutionswerereviewed:The Johnson City Medical Center, a tertiary care fa-cility, and the Veterans Affairs Medical Center. Bothof these hospitals are affiliated with our university.Six hundred seventy-nine patients were treated withGP IIb/IIIa inhibitors but did not undergo a percuta-neous coronary intervention (PCI) during their admis-sion. The search was done electronically based on thecurrent procedural terminology (CPT) codes for AMI,AMI and PCI, AMI and AH, AMI and hemoptysis,PCI and AH, and PCI and hemoptysis. All patientswho were treated with either eptifibatide or abciximabbetweenAugust1998andJanuary15,2005,wereiden-tified through the electronic records of the in-patientpharmacy at our institute. The records of 4,136 pa-tients admitted with AMI but not receiving GP IIb/IIIainhibitors were similarly evaluated. This study was ap-proved by our institutional review board and researchand development department.A total of 52 patients (35 in the treated group and 17inthecontrolgroup)wereidentifiedashavingsustainedAH during the hospitalization by the CRT coding.(Fig.1).Therecordsofthesepatientswerereviewedbyhand and all patients who did not fulfill the inclusioncriteria were excluded. We defined AH as an episodeofspontaneoushemoptysisassociatedwithatleastoneof the following: new infiltrates on radiological exam-ination (chest X-ray, CT scan), a drop in hemoglobinof at least 2 g/dL not explained by any extrapulmonaryhemorrhagic process, or a drop in oxygen saturationunlessrelatedtoapulmonaryprocessclearlyunrelatedtoAH.Ifthepatientdidnothavehemoptysis,twooftheothercriterianeededtobefulfilled.AHandmyocardialinfarctionhadtohaveoccurredduringthesameadmis-sion for patients to be included in the study. Wheneveravailable, bronchoscopic data were used to confirm orrejectthediagnosisofdiffuseAH.Fortypatientsdidnotfulfillourinclusioncriteriaandwereexcluded.Twenty-fourpatientsinthetreatedgroupwereexcluded(12fornot having the AH on the pertinent admission, 10 forhaving pneumonia and hemoptysis, and two for havingthe episode of hemoptysis prior to the administrationof the GP IIb/IIIa inhibitor). Sixteen patients were ex-cluded from the control group (10 for not having theAHonthepertinentadmissionandsixforhavingpneu-monia and hemoptysis). Twelve patients (11 of whomwere on a GP IIb/IIIa inhibitor and one who was in thecontrol group) fulfilled our criteria.Allrecordswerereviewedfordemographicandclin-ical characteristics including symptoms, timing anddose of eptifibatide, and abciximab, time to onset of bleeding, previous history of pulmonary disease, andlaboratory data including chest radiographic changes, Vol. 19, No. 4, 2006 Journal of Interventional Cardiology 357  ISKANDAR, ET AL.  9594 Acute Myocardial Infarction 5458 (56%) GP IIB/IIIA I  NO GP IIB/IIIA I 4136 (44%) 17 AH 16 met exclusion criteria: 10 had AH prior toadmission6 had pneumonia 1 patient 0.02% 3648 Abciximab 1810 Eptifibatide 33% 67% 35 AH 11 patients 0.2% 6 Eptifibatide 5 Abciximab 0.33% 0.14% GP IIB/IIIA I: glycoprotein IIB/IIIA inhibitorsAH: alveolar hemorrhage 24 met exclusion criteria: 12 had AH prior to admission 10 had pneumonia 2 had hemoptysis prior to GPIIB/IIIA I Figure 1.  Algorithm for expanded inclusion and exclusion criteria of all patients. peak activated clotting time (ACT), PT, PTT, INR,platelet count, hemoglobin, hematocrit, blood urea ni-trogen(BUN),andcreatinine.Allantiplateletandanti-coagulatoryagentswerelistedforeachpatient.Clinicaloutcome at discharge was recorded. Statistical Analysis.  Data were analyzed usingSAS version 9.2 statistical software. Categorical vari-ables were compared by the exact chi-square test andFisher’s exact test. A P value of  < 0.05 was consideredsignificant. Results Abciximab and eptifibatide were used in 3,648(67%) and 1,810 (33%) patients, respectively (Fig-ure 1). Six patients out of 1,810 patients (0.33%) whoreceivedeptifibatide(fivemales;meanage,68.5years;rangefrom47to81years)developedAH.Fivepatientsout of the 3,648 patients (0.14%) treated with abcix-imab (two men; mean age, 68.4 years; range, 58–79years) developed AH. Their demographics and clinicaland laboratory data are detailed in Tables 1 and 2.Eleven GP IIb/IIIa-treated patients had AH (0.2%),comparedtoonlyoneintheuntreatedgroups(0.025%).None of these patients had previously received GPIIb/IIIa blockers. The incidence of AH in GP IIb/IIIainhibitor-treated patients differed significantly fromthat of untreated patients (chi-square, 5.92; Fisher’sexact test, P = 0.017) (Table 3). No significant differ-ences were found between AH patients treated witheptifibatide and abciximab (chi-square, 2.27; Fisher’sexact test, P = 0.19). However, comparing each drugto the control group, eptifibatide showed a statisti-callysignificantdifferenceintheincidenceofAH(chi-square, 10.1; Fisher’s exact test, P = 0.004) while ab-ciximab did not (chi-square, 3.2; Fisher’s exact test,P = 0.105).All patients had received a standard dose (per phar-macy protocol) of eptifibatide (180 µ g/kg; maximum:22.6mg),followedbyacontinuousinfusionatarateof 2 µ g/kgperminute(maximum:15mg/h))orabciximab(bolus of 0.25 mg/kg) followed immediately by an in-fusion at a rate of 1.25 µ g/kg per minute (maximum of 10  µ g/kg for 12 hours). When necessary, the dose of eptifibatide was adjusted according to creatinine clear-ance. No patient had a prior history of bleeding. Noneof the patients were on warfarin.The mean peak ACT was 255 seconds (240 and270 seconds) in the eptifibatide group and 258 sec-onds(range,220–292seconds)intheabciximabgroup.Only two patients in the eptifibatide group receivedblood transfusions (2 and 4 units). The LVEDP waselevated in four of the five eptifibatide-treated pa-tients (80%) (Mean: 27.4, range 11–37) and in three of the five abciximab-treated patients (60%) (Mean: 23,range 14–33). Nine of the 11 patients were smokers orex-smokers. One patient in the control group had AH . Thispatient received thrombolytic therapy with Retavaseandunderwentcardiaccatheterizationwithsubsequent 358 Journal of Interventional Cardiology Vol. 19, No. 4, 2006  ALVEOLAR HEMORRHAGE AND GP IIB/IIIA INHIBITORS Table 1.  Clinical Variables in Patients Developing Pulmonary Hemorrhage on EptifibatideAge/Sex 81F 76M 72M 66M 47M 69MWeight (kg) 80 71 62.7 113 88.6 113Time to Dx 72 h 48 h 5 h 24 h 32 h 12 hLab data Hb (before/after) 13.5/10.9 11.6/11.4 10.9/8.8 11.7/8.8 14.8/11.2 10/9.5Plat 208 159 84 263 95 142INR 1.3 1.2 1.1 1.36 1.24 1.37ACT 270 NA 240 NA NA NAPTT 58 31 46.7 53.2 32 30BUN/Cr 19/1.6 38/1.7 34/0.9 25/1.8 14/1.0 17/1.0PaO 2  65 106 58 59 68 56LV-EDP 11 NA 30 37 25 34Blood Products  ∼ ∼  2U PRBCs 4U PRBCs  ∼ ∼ Dx Hemoptysis Hemoptysis Hemoptysis Hemoptysis Hemoptysis HemoptysisPlavix 300 mg 75 mg 75 mg 75 mg 75 mg NASA Y Y Y Y Y YHeparin Y LMWH Y Y Y LMWHLytics N N Reteplase N N NCOPD N Y N N N YSmoker N Ex Y Ex Y ExCXRay BLLI BLLI BLLI RLLI NL LLIFull recovery from AH Y Y Y Y Y YF = female; M = male; D = day; h = hour; kg = kilograms; Plat = platelet count; INR = international normalization ratio; ACT = activatedclotting time; PTT = partial thromboblastin time; BUN = blood urea nitrogen; Cr = creatinine; PaO 2  = arterial oxygen pressure; LV-EDP = left ventricular end-diastolic volume; ASA = aspirin; Lytics = thrombolytics; COPD = chronic obstructive lung disease; CXRay = chest film;AH = alveolar hemorrhage; Y = yes; N = no; NA = not available; LLLI = left lower lobe infiltrate; BLLI = bilateral lower lobe infiltrate;LMWH = low molecular weight heparin; Ex = ex-smoker; U PRBCs = units of packed red blood cells. emergent coronary artery bypass graft (CABG). HisLVEDP was elevated.Signs of AH including hemoptysis, abnormal chestradiographic finding, drop in hemoglobin, or arterialhypoxemia appeared within 5–72 hours after the firstdose of abciximab or eptifibatide. Hemoptysis was in-variably present. Three of these patients had a normalinitial chest X-ray (one in the eptifibatide group andtwo in the abciximab group). All patients had hypox-emia, but only seven of 11 (63%) exhibited a dropin hemoglobin (2–3.6 g/dL). There was no death at-tributable to AH. Conservative treatment withholdingGP IIb/IIIa drug therapy resulted in uneventful recov-ery in all patients. Discussion AH associated with GP IIb/IIIa inhibitor is rare, butpotentially life-threatening. 11 It may be underrecog-nized because radiographic features of AH and cardio-genic pulmonary edema fail to exhibit diagnosticallyuseful differences. 12 The first case of AH associatedwith the use of abciximab was reported by Sitges etal. 13 Subsequently, Ali et al. described AH in a pa-tient receiving tirofiban. 14 Khanlou et al. reported sixadditional cases of abciximab-induced AH. 15 , 16 Onlyone case of alveolar bleeding has been reported in pa-tients receiving eptifibatide. 17 By December 1998, Co-hen and Effron were able to collect 109 reports of pul-monarybleedingaftertreatmentwithabciximab. 15 Stillonly 10 of the 5,382 patients (0.19%) receiving abcix-imab in four clinical trials developed AH. 15 AH, compared with other hemorrhagic syndromes,is difficult to diagnose and may be easily overlooked.Presenting signs may include hemoptysis, acute res-piratory distress syndrome, and new pulmonary infil-trates by radiological examination. These symptomsand signs are quite nonspecific, making the diagnosisof AH dependent upon a high degree of clinical suspi-cion. 18 , 19 Previousreportsofthefrequencyandseverityof AH after abciximab may therefore lack in reliabilityand may have failed to distinguish AH from other syn-dromessuchaspneumoniaandcongestiveheartfailureexacerbation. In addition, previous reports failed to in-clude appropriate untreated control groups. Mortalityfrom AH in patients treated with GP IIb/IIIa inhibitorshas been variable, ranging between 0% (this report) to28%. 8 , 10 , 20 Vol. 19, No. 4, 2006 Journal of Interventional Cardiology 359  ISKANDAR, ET AL. Table 2.  Clinical Variables in Patients Developing Pulmonary Hemorrhage on AbciximabAge/Sex 78F 79F 58F 65M 62MWeight (kg) 83 96 73 57 95Time to Dx 3d  < 12 h NA  < 24 h  < 6 hLab data Hb (before/ after) 13/9.4 10.9/10.4 12.9/10.6 16.1/13.2 16.6/16.8Plat 190 164 214 126 127INR NA 1.1 0.9 1.1 0.8ACT 220 291 292 NA 227PTT NA 54 58 30 178BUN/Cr 18/1.2 25/1.4 17/0.8 25/1.1 12/0.9PaO 2  NA NA NA 48 64LV-EDP 26 18 Emergent LHC 14 33Dx Hemoptysis Hemoptysis Hemoptysis Hemoptysis HemoptysisPlavix 150 mg 150 mg N N 300 mgASA Y Y Y Y YHeparin LMWH N Y N YLytics Retavase N N N NCOPD N N Y  ∼  YSmoker Y N Y Y YCXRay LLLI Normal LLLI BLLI NormalFull recovery from AH Y Y Y Y YF = female; M = male; d = day; h = hour; kg = kilograms; Plat = platelet count; INR = international normalization ratio; ACT = activatedclotting time; PTT = partial thromboblastin time; BUN = blood urea nitrogen; Cr = creatinine; PaO 2  = arterial oxygen pressure; LV-EDP = left ventricular end-diastolic volume; ASA = aspirin; Lytics = thrombolytics; COPD = chronic obstructive lung disease; CXRay = chest film;AH = alveolar hemorrhage; Y = yes; N = no; NA = not available; LLLI = left lower lobe infiltrate; BLLI = bilateral lower lobe infiltrate;LMWH = low molecular weight heparin; LHC = left heart catheterization. Cases of AH have been reported sporadically for allthree GP IIb/IIIa inhibitors. 13–15 , 21–24 In a study by Aliet al., the incidence of pulmonary hemorrhage for ab-ciximab, eptifibatide, and tirofiban were 0.7%, 0.5%,and 0.9%, respectively (P = NS). 20 A prolonged ACTas well as an elevated PTT was present in most pa-tients. 20 AH appears to have occurred less frequently Table 3.  Incidence of AH in Treatment subgroupsOR CI Chi-square P ValueAll GP IIb/IIIa 8.35 (1–359) 5.92 0.017 ∗ I vs controlEptifibatide 2.42 (0.6–10) 2.27 0.19 † vs AbciximabEptifibatide 13.7 (1.6–632) 10.1 0.004 ∗ vs controlAbciximab 5.6 (0.6–268) 3.2 0.105 † vs controlAH  =  alveolar hemorrhage. OR  =  Odds ratio. CI  =  95 percentconfidence interval. P value  =  P value according to Fisher’s exacttest. GP IIb/IIIa I = glycoprotein IIb/IIIa inhibitors. ∗ Indicates statistical significance (P < 0.05). † Did not reach statistical significance. in the clinical trials of abciximab with only 10 of the5,382 patients (0.19%) receiving the diagnosis. Fre-quencies of AH appeared high in an early trial in-volvingthecoadministrationofhighdosedheparin 4 , 8 , 6 and exceeded 0.3% in one of the trials. 5 However,in later trials, using weight-adjusted heparin doses,lower incidence rates were observed (0.04%). 21 , 5 Onthe other hand, pulmonary bleeding has been onlysparsely reported with eptifibatide treatment. Use of weight-adjusted heparin to maintain an ACT of 200–250 U during coronary intervention has been shownto reduce the incidence of major bleeding. 5 In our pa-tients, the ACT ranged from 220 to 292 with a meanof 257 U. Rigorous attention to heparin dosing duringadministrationoftheseagentswaslikelythereasonforthe low fatal (0% mortality) and nonfatal complicationrate in our series.AH is characterized by widespread bleeding fromthe intrapulmonary microvasculature. Pericapillary in-flammatory cell infiltrates appear to be a commonhistopathologic finding especially in patients suffer-ing from inflammatory vasculopathies such as colla-gen vascular disease, 25 but no relation has been foundbetween antiplatelet agents and vasculitis. Capillaritis 360 Journal of Interventional Cardiology Vol. 19, No. 4, 2006
Similar documents
View more...
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks