meta-analysis low-molecular-weight heparin and bleeding in patients.pdf

Meta-Analysis: Low-Molecular-Weight Heparin and Bleeding in Patientswith Severe Renal InsufficiencyWendy Lim, MD, BSc; Francesco Dentali, MD; John W. Eikelboom, MBBS; and Mark A. Crowther, MD, MSc

Background: Dose adjustment or laboratory monitoring of low-molecular-weight heparin (LMWH) is commonly recommended forpatients with severe renal insufficiency (creatinine clearance �30mL/min), but the basis for this recommendation is unclear.

Purpose: To compare levels of anti-Xa heparin and risk for majorbleeding in LMWH-treated patients with a creatinine clearance of30 mL/min or less versus those with a creatinine clearance greaterthan 30 mL/min by using standard weight–adjusted therapeuticdoses, empirically adjusted doses, or prophylactic doses of LMWH.

Data Sources: Electronic databases (MEDLINE, EMBASE, and theCochrane Library) searched to December 2005 with no languagerestrictions. The authors also searched reference lists and contactedexperts.

Study Selection: Observational or subgroups of randomized studiesthat included non–dialysis-dependent patients with varying degreesof renal function who were treated with LMWH and reportedcreatinine clearance and anti-Xa levels or major bleeding.

Data Extraction: Two reviewers independently selected studies andextracted data on patient characteristics, renal function, LMWHtreatment, anti-Xa levels, and major bleeding. The pooled oddsratio of major bleeding in patients with a creatinine clearance of 30mL/min or less was calculated by using the Peto method.

Data Synthesis: Eighteen studies using 3 preparations of LMWH(15 studies using enoxaparin, 2 using tinzaparin, and 1 using dalte-parin) were included. Peak anti-Xa levels measured 4 hours after asubcutaneous injection were statistically significantly higher in pa-tients with a creatinine clearance of 30 mL/min or less comparedwith those with a creatinine clearance greater than 30 mL/min in

studies that used a standard therapeutic dose of enoxaparin (4studies) but not in studies of empirically dose-adjusted enoxaparin(3 studies). Data were insufficient to assess the relationship be-tween anti-Xa and renal function for prophylactic doses of enox-aparin and therapeutic doses of tinzaparin or dalteparin. In 12studies involving 4971 patients, LMWH was associated with astatistically significant increase in the risk for major bleeding inpatients with a creatinine clearance of 30 mL/min or less comparedwith those with a creatinine clearance greater than 30 mL/min(5.0% vs. 2.4%; odds ratio, 2.25 [95% CI, 1.19 to 4.27]; P �0.013). When analyzed according to LMWH preparation, majorbleeding was increased when a standard therapeutic dose of enox-aparin was used (8.3% vs. 2.4%; odds ratio, 3.88 [CI, 1.78 to8.45]) but may not be increased when an empirically adjusted doseof enoxaparin is used (0.9% vs. 1.9%; odds ratio, 0.58 [CI, 0.09 to3.78]; P � 0.23 for heterogeneity). There were insufficient studiesto assess the risk for major bleeding with tinzaparin, dalteparin, andprophylactic doses of enoxaparin.

Limitations: The data for tinzaparin and dalteparin were limited.Data are observational, and the potential for confounding cannotbe excluded.

Conclusions: Non–dialysis-dependent patients with a creatinineclearance of 30 mL/min or less who are treated with standardtherapeutic doses of enoxaparin have elevated levels of anti-Xa andan increased risk for major bleeding. Empirical dose adjustment ofenoxaparin may reduce the risk for bleeding and merits additionalevaluation. No conclusions can be made regarding other LMWHs.

Ann Intern Med. 2006;144:673-684. www.annals.orgFor author affiliations, see end of text.

One of the most important advantages of low-molecu-lar-weight heparin (LMWH) compared with unfrac-

tionated heparin (UFH) is its predictable anticoagulantresponse, which allows it to be administered in fixed,weight-based doses without laboratory monitoring. Whenused to treat patients with acute venous thromboembolism,LMWH is associated with lower rates of recurrent throm-bosis, bleeding, and death than is weight-adjusted, moni-tored UFH (1, 2). These advantages have revolutionizedmanagement of acute venous thromboembolism by allow-ing most patients to be treated out of hospital. Low-mo-lecular-weight heparin has also greatly simplified in-hospi-tal management of selected patients with venousthromboembolism and those with acute coronary syn-dromes. However, uncertainty still surrounds use ofLMWH in patients with severe renal insufficiency becauseit is excreted by the kidneys (3) and, unlike UFH, its an-ticoagulant effect cannot be completely reversed. Althoughmost randomized trials of LMWH excluded patients withrenal insufficiency (4), pharmacokinetic studies suggest an

association between creatinine clearance and levels of anti-factor Xa heparin (which measures the anticoagulant effectof LMWH), and increased bleeding complications havebeen reported when LMWH is used in patients withchronic renal insufficiency (5, 6).

Because of concerns about the risk for accumulationand bleeding in patients with renal impairment, the Amer-ican College of Chest Physicians and the College of Amer-ican Pathologists recommend UFH instead of LMWH in

See also:

PrintEditors’ Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 674

Web-OnlyAppendixCME quizConversion of figures and tables into slides

Annals of Internal Medicine Review

© 2006 American College of Physicians 673

Downloaded From: http://annals.org/ by a Univeristy of Texas User on 01/18/2015

patients with a creatinine clearance of 30 mL/min or less,or monitoring of anti-Xa activity if LMWH is used (4, 7).However, the evidence for this recommendation is con-flicting (8) and there are no reliable data to guide theinterpretation of anti-Xa levels to monitor treatment or toadjust LMWH doses (7). Despite this limitation, anti-Xalevels are the only available method to monitor LMWHactivity and their use in clinical practice is based on con-sensus recommendations. Peak anti-Xa levels occur 4 hoursafter a therapeutic dose of subcutaneous LMWH is admin-istered. Peak levels above the upper limit of the recom-mended therapeutic range (0.6 to 1.0 IU/mL) may be as-sociated with an increased risk for bleeding (4).

To further clarify the relationship between LMWHanti-Xa levels and creatinine clearance and the risk forbleeding in patients with a creatinine clearance of 30 mL/min or less, we performed a systematic review of all studiesof LMWH in non–dialysis-dependent patients with vary-ing degrees of renal function that reported creatinine clear-ance and anti-Xa levels or major bleeding events.

METHODS

This study was conducted according to the Meta-anal-ysis Of Observational Studies in Epidemiology (MOOSE)guidelines (9).

Study IdentificationWe attempted to identify all published and unpub-

lished studies of LMWH in non–dialysis-dependent pa-tients with renal insufficiency using the MEDLINE (1966to week 3 of November 2005), EMBASE (1980 to 2005,week 51), and Cochrane Library (2005, Issue 4) electronic

databases. The search was completed on 19 December2005. The search strategy, which had no language restric-tions, was developed in collaboration with a professionallibrarian (Appendix, available at www.annals.org). Wemanually reviewed the reference lists of all retrieved articlesand contacted content experts for additional published orunpublished trials.

Study SelectionStudy selection was performed independently and in

duplicate; disagreements were resolved through discussion.Studies were included if they 1) had at least 10 patients; 2)involved patients with varying degrees of renal function,including those with a creatinine clearance of 30 mL/minor less; 3) administered at least 1 dose of a commerciallyavailable LMWH; and 4) reported at least 1 of the follow-ing outcomes: anti-Xa levels or major bleeding. Studieswere excluded if patients were dialysis-dependent, ifLMWH was administered intravenously, if more than 1LMWH preparation or dose (for example, prophylacticand therapeutic doses) were used, and if the data for an-ti-Xa or bleeding for individual LMWH preparations ordoses could not be separately analyzed.

Study Quality AssessmentTwo unmasked investigators independently assessed

study quality according to the following criteria: 1) methodof patient enrollment—studies with consecutive patientenrollment were considered to be of higher quality thanthose with nonconsecutive enrollment; 2) duration andcompleteness of patient follow-up for major bleeding—studies with longer and more thorough follow-up wereconsidered to be of higher quality because major bleedingoutcomes are more accurately reported; and 3) method ofoutcome ascertainment—studies in which bleeding wasobjectively and reproducibly assessed by using a priori def-initions were rated higher than those that did not usebleeding definitions or those that had unclear or nonobjec-tive methods of measurement.

Data ExtractionTwo unmasked investigators independently extracted

data on study design, patient characteristics, renal func-tion, type and dose of LMWH, anti-Xa levels, and numberof major bleeding episodes. Data were extracted in dupli-cate, and disagreements were resolved through discussion.We defined a creatinine clearance of 30 mL/min or less assevere renal insufficiency, calculated by using the Cock-croft–Gault (10) or Modification of Diet in Renal Disease(MDRD) (11) equation. For studies that adopted otherthresholds to define severe renal failure or those that re-ported only mean creatinine clearance, we contacted theauthors to obtain the data corresponding to a creatinineclearance threshold of 30 mL/min.

We extracted peak anti-Xa levels 4 hours after subcu-taneous injection of LMWH because peak levels correlatemore strongly with efficacy and safety than do trough levels(12). Because maximum LMWH activity (Amax) is ob-

Context

The risks of low-molecular-weight heparin (LMWH) in pa-tients with severe renal insufficiency are not clear.

Contribution

In this review of 12 studies, patients with severe renal in-sufficiency receiving LMWH had an increased risk for ma-jor bleeding events. Four studies found that fixed-doseenoxaparin had greater anticoagulant effects in these pa-tients. Three studies suggested that empirically dose-ad-justed enoxaparin might not increase anticoagulant effects.

Cautions

Evidence relating to LMWHs other than enoxaparin wasscant.

Implications

Patients with severe renal insufficiency receiving standardfixed-dose enoxaparin have greater anticoagulant effectsand a higher risk for major bleeding. Empirical dose adjust-ment may reduce the risk for such events.

—The Editors

Review Low-Molecular-Weight Heparin in Renal Insufficiency

674 2 May 2006 Annals of Internal Medicine Volume 144 • Number 9 www.annals.org


served at 4 hours, Amax data were considered equivalent to4-hour anti-Xa levels. A meta-analysis of anti-Xa levels at 4hours was not performed because measures of variancewere not reported in most studies, thereby precludingpooling of the data. We report the anti-Xa data accordingto LMWH type because differences in molecular weightand charge density of LMWH may result in differences inrenal clearance and anti-Xa levels (13, 14). The LMWHs,such as tinzaparin, that are structurally similar to UFHmay be less prone to accumulate in patients with re-nal insufficiency than smaller, less negatively chargedLMWHs, such as enoxaparin.

We accepted the definitions of major bleeding re-ported by authors and did not reclassify events. Therapeu-tic doses of LMWH (enoxaparin, 1.5 mg/kg of bodyweight once daily or 1 mg/kg twice daily; tinzaparin, 175IU/kg once daily; dalteparin, 200 IU/kg once daily or 100IU/kg twice daily) and prophylactic doses (enoxaparin, 40mg once daily or 30 mg twice daily) were defined accord-ing to the product monographs. Adjusted-dose LMWHwas defined as any dose that was modified empirically oraccording to renal function or measured anti-Xa levels.

If the required data could not be extracted from thepublished report, we contacted the corresponding authorby e-mail to request additional data. If a response was notreceived after 15 days, we sent a second e-mail; if we stillreceived no response, we contacted secondary authors.

Statistical AnalysisTo assess the agreement between reviewers for study

selection, we used the k statistic, which measures chance-corrected agreement (15). The odds ratio for major bleed-ing in patients with and without severe renal insufficiencytreated with LMWH was calculated for each study andpooled with the Peto method (fixed-effects) by using Re-view Manager, version 4.2 (RevMan, Oxford, UnitedKingdom; the Cochrane Collaboration, 2003). Statisticalheterogeneity was evaluated by using the I2 statistic, whichmeasures the extent of inconsistency among study resultsand describes the proportion of total variation in studyestimates that is due to heterogeneity rather than samplingerror. We preferred the Peto method to other statisticalapproaches for combining odds ratios because it providesrelatively unbiased estimates of treatment effects when theevent rates are low. We also calculated a pooled risk differ-ence and 95% CIs by using exact statistical methods forstratified 2 � 2 tables.

Two a priori secondary analyses were performed. Thefirst compared major bleeding according to LMWH typebecause accumulation may vary among different prepara-tions in patients with renal insufficiency (13, 14). Thesecond compared major bleeding according to LMWHdose. Because bleeding is correlated with the intensity ofanticoagulation, studies using therapeutic doses reportedhigher bleeding rates than those using prophylactic doses.

A sensitivity analysis that included only high-quality

studies was performed to assess the robustness of our pri-mary results. A funnel plot of effect size versus standarderror was used to assess possible publication bias (16).

Role of the Funding SourceNo funding was received for this study.

RESULTS

Study Identification and SelectionOur search identified 875 published studies (219 from

MEDLINE, 792 from EMBASE, and 24 from theCochrane Library); 141 studies were duplicated in the 3databases (Figure 1). Contact with experts did not revealadditional published or unpublished studies. We excluded839 studies after screening their titles and abstracts andretrieved the remaining 36 studies for more detailed eval-uation (17–52). Of the 36 retrieved studies, 19 were ex-cluded: 3 contained duplicate data (21, 32, 42); 2 did notinclude patients with a creatinine clearance of 30 mL/minor less (25, 44); 1 only included patients with a creatinineclearance of 30 mL/min or less (20); 3 involved hemodialysis-dependent patients (23, 41, 48); 1 used a noncommerciallyavailable LMWH (47); 3 administered LMWHs intrave-nously (35, 46, 52); 3 used therapeutic- and prophylactic-dose LMWH and separate data for the 2 groups could notbe obtained (33, 40, 43); 2 examined LMWH populationpharmacokinetics but data for patients with severe renalinsufficiency could not be obtained (22, 51); and 1 registryused different LMWH preparations and data specific forLMWH type could not be obtained (50). One additionalstudy was identified through manual review of references(53). Thus, 18 studies were included in the systematic re-view (17–19, 24, 26–31, 34, 36–39, 45, 49, 53), 13 thatreported anti-Xa levels (17, 19, 26, 30, 31, 34, 36–39, 45,49, 53) and 12 that compared the rate of major bleeding inpatients with and without renal insufficiency (17–19, 24,26–30, 36, 39, 53). The interobserver agreement for studyselection was excellent (k � 0.98).

Study CharacteristicsAll studies were published in English and included 18

to 3501 patients (Table 1). Two studies reported resultsfrom subgroups of patients enrolled in randomized trials ofenoxaparin for acute coronary syndromes (29, 37), 12 wereprospective cohort studies (17, 18, 26, 30, 31, 34, 36, 38,39, 45, 49, 53), and 4 were retrospective cohort studies(19, 24, 27, 28).

Creatinine clearance was calculated by using the Cock-croft–Gault equation in 12 studies (17–19, 24, 26, 28, 29,34, 36, 37, 49, 53) and by using the Modification of Dietin Renal Disease (MDRD) equation in 2 studies (27, 30).One study measured endogenous creatinine clearance (38),1 used the Kampmann–Siersbaek-Nielsen equation (45),and 2 did not specify the method of calculation (31, 39).

The indications for LMWH treatment included acutecoronary syndromes, atrial fibrillation, venous thromboem-

ReviewLow-Molecular-Weight Heparin in Renal Insufficiency

www.annals.org 2 May 2006 Annals of Internal Medicine Volume 144 • Number 9 675


bolism, and prevention of thromboembolic complicationsin patients with prosthetic heart valves. Two pharmacoki-netic studies evaluated LMWH in healthy persons or inpatients with renal insufficiency (38, 45).

Fifteen studies used enoxaparin (17–19, 24, 26–31,34, 37–39, 45), 2 used tinzaparin (36, 53), and 1 useddalteparin (49). Of the 15 enoxaparin studies, 7 used ther-apeutic doses, 4 used empirically adjusted doses, 3 usedprophylactic doses, and 1 used both therapeutic and pro-phylactic doses. The 2 tinzaparin studies and the dalteparinstudy used therapeutic doses. Fourteen of the 16 enoxapa-rin studies and the tinzaparin and dalteparin studies ad-

ministered multiple doses. All studies measuring anti-Xalevels used chromogenic anti-Xa assays.

Study QualityConsecutive patients were enrolled in 12 of the 14

prospective studies; enrollment was nonconsecutive in the2 pharmacokinetic studies (38, 45). Mean duration of fol-low-up for bleeding events ranged from 5 to 45 days, andmost studies followed patients during their hospital staybut did not report the actual duration of follow-up. Fol-low-up was not evaluable for retrospective studies or thosemeasuring anti-Xa levels only. Major bleeding was defineda priori in 10 of the 12 studies that reported bleedingoutcomes (17, 19, 24, 27–30, 36, 39, 53); 2 of these stud-ies adapted the Thrombolysis in Myocardial Infarction(TIMI) criteria for bleeding (19, 39), 1 used a modifiedversion of the Warfarin Optimized Outpatient Follow-upStudy Classification (17), and 7 defined major bleedingusing 1 or more of the following criteria: decrease in he-moglobin level, transfusion requirements, bleeding into acritical organ, or death. None of the prospective studiesreported the frequency of bleeding assessments.

Measured OutcomesAnti-Xa Measurements

Enoxaparin. Ten enoxaparin studies measured anti-Xalevels: 4 therapeutic-dose studies (17, 26, 30, 37), 3 adjust-ed-dose studies (19, 31, 39), and 3 prophylactic-dose stud-ies (34, 38, 45) (Table 2). For the 4 therapeutic-dose stud-ies using 1 mg/kg every 12 hours, mean anti-Xa levels aftera minimum of 3 doses ranged from 1.27 to 1.58 IU/mL inpatients with a creatinine clearance of 30 mL/min or lessand 0.91 to 1.06 IU/mL in patients with a creatinine clear-ance greater than 30 mL/min. In 3 of the 4 studies, levelsof anti-Xa were statistically significantly higher among pa-tients with a creatinine clearance of 30 mL/min or less (26,30, 37).

Three studies evaluated empirically adjusted doses ofenoxaparin. In 2 studies, patients with a creatinine clear-ance of 30 mL/min or less received 65% of the recom-mended enoxaparin dose of 1 mg/kg every 12 hours for thefirst 3 doses; subsequent doses were adjusted according tothe peak anti-Xa level measured 4 hours after the thirdinjection, aiming for a target anti-Xa level of 1.0 IU/mL(31, 39). One of these studies compared the peak anti-Xalevel in patients with and without renal insufficiency andfound no statistically significant difference (31). In thethird study, all patients received an initial dose of 1 mg/kg,but patients with a creatinine clearance of 30 mL/min orless subsequently received 0.5 mg/kg and those with a cre-atinine clearance of 30 to 60 mL/min received 0.75 mg/kg,both administered every 12 hours (19). The resulting anti-Xa levels remained within the predefined therapeutic tar-get levels of anti-Xa specified by the authors (0.60 to1.00 IU/mL); the mean anti-Xa level was 0.65 IU/mL

Figure 1. Study selection progression.

LMWH � low-molecular-weight heparin.




Table 1. Study and Patient Characteristics*

Study, Year(Reference)

Study Designand Method ofEnrollment

FundingSource

Indication forLMWH

CreatinineClearance,mL/min†

Patients,n

Mean Dose MajorBleeding,%‡

Bleeding Follow-up§:Mean Duration;Completeness

EnoxaparinTherapeutic dose

Bazinet et al.,2005 (17)

Prospectivecohort; C

Industry ACS, AF, VTE 10–3031–50�50

3654

106

1 mg/kg every 12 hor 1.5 mg/kg per d

1 (2.8)2 (3.7)0

5 d;92% follow-up

for anti-XaMacie et al.,

2004 (24)Retrospective

cohortNot funded ACS �25

�257

2011 mg/kg every 12 h 2 (28.6)

6 (3.0)In-hospital (10 d);

NAPeng et al.,

2004 (26)Prospective

cohort; CProfessional

organizationACS, AF, VTE,

prosthetic valves�3031–60�60

71924

1 mg/kg every 12 h 0 In-hospital (NS);no loss tofollow-up

Thorevska et al.,2004 (27)

Retrospectivecohort

NS ACS, AF, VTE,other

�30�30

65171

1 mg/kg every 12 h 7 (10.8)11 (6.4)

In-hospital (NS); NA

Spinler et al.,2003 (29)

Randomized,controlled trialsubgroup; C

Industry ACS �30�30

693432

30 mg IV loading dose,then 1 mg/kg every12 h

5 (7.5)74 (1.1)

ESSENCE: 30 d,TIMI 11B: 72 h,in-hospital and day43; NS

Chow et al.,2003 (30)


NS ACS, AF, VTE,stroke

�1011–3031–60�60

1467

1 mg/kg every 12 h 0000

In-hospital (NS);no loss tofollow-up

Khazan et al.,2003 (28)

Retrospectivecohort

Hospital AF �30�30

1761

1 mg/kg every 12 hor 1.5 mg/kg per d

2 (11.8)3 (4.9)

While receivingenoxaparin(NS); NA

Becker et al.,2002 (37)

Randomized,controlled trialsubgroup;C; NS

Industry ACS �4040–80�80

11149273

30 mg IV loading dose,then 1 mg/kg every12 h or 1.25 mg/kgevery 12 h

Unable toobtain

14 d; NS

Prophylactic doseKhazan et al.,

2003 (28)Retrospective

cohortNS AF �30

�303647

30 mg every 12 h or40 mg/d

3 (8.3)3 (6.4)


Mahe et al.,2002 (34)


NS VTE prophylaxis �40�40

464

4000 IU (40 mg) daily Notassessed

–

Sanderink et al.,2002 (38)

Prospectivecohort; NC

Industry PK study involunteers

�3031–5051–80�80

12121212

40 mg/d Notassessed

–

Cadroy et al.,1991 (45)

Prospectivecohort; NC

Industry PK study involunteers

�20“Normal”

1212

0.5 mg/kg single dose Notassessed

–

Adjusted doseGreen et al.,


cohort; CIndustry ACS �30

30–5051–80�80

181163

Creatinine clearance�25 mL/min: 0.75mg/kg every 12 h;creatinine clearance�25 mL/min: 1mg/kg every 12 h

1 (5.6)000

NS

Kruse and Lee,2004 (19)

Retrospectivecohort

NS ACS, AF, VTE,other

�30

30–60

50

120

1 mg/kg every 12 h(initial dose), then 0.5mg/kg every 12 h

0.75 mg/kg every12 h

0

1 (0.8)

In-hospital (NS); NA

Khazan et al.,2003 (28)

Retrospectivecohort

NS AF �30�30

1042

Not specified 03 (7.1)


Collet et al.,2003 (31)


Professionalorganization

ACS �30�30

62453

0.70 (SD, 0.07) mg/kgevery 12 h

Unable toobtain

30 d; NS (88.5%follow-up forischemic events)



NS ACS �3030–60�60

282855

0.64 (SD, 0.04) mg/kgevery 12 h, 0.84 (SD,0.03) mg/kg every 12h, 0.92 (SD, 0.03)mg/kg every 12 h

01 (3.6)0

In-hospital (NS); noloss to follow-up

TinzaparinPautas et al.,


cohort; CNot funded ACS, AF, VTE,

other20–3435–4950–64�65

51604445

175 IU/kg daily 01 (1.7)02 (4.4)

30 d; no loss tofollow-up

Continued on following page




(mean, 0.19) for patients with a creatinine clearance of 30mL/min or less and 0.82 IU/mL (mean, 0.18) for patientswith a creatinine clearance of 30 to 60 mL/min (P �0.001).

Two of the prophylactic-dose studies evaluated multi-ple doses using the recommended dose of 40 mg daily (54),and 1 evaluated a single dose of 0.5 mg/kg. With the40-mg dose, 1 study showed no difference in the mean4-hour anti-Xa level in patients with a creatinine clearanceless than 40 mL/min versus those with a creatinine clear-ance of 40 mL/min or greater (34); a pharmacokineticstudy found that mean anti-Xa levels were almost 40%higher in patients with a creatinine clearance of 30 mL/min or less versus those with a creatinine clearance greaterthan 30 mL/min (38). A single dose of 0.5 mg/kg resultedin higher levels of anti-Xa in patients with a creatinineclearance of 20 mL/min or less than in those with normalrenal function (creatinine clearance �85 mL/min) (45).However, in each study, peak levels of anti-Xa remainedbelow the lower limit of the usual target therapeutic rangefor LMWH.

Tinzaparin. Two therapeutic-dose studies evaluatedlevels of anti-Xa in elderly patients with impaired renalfunction (36, 53). These studies did not find a correlationbetween the peak level of anti-Xa and creatinine clearance,but comparison of levels of anti-Xa between patient groupswas not reported.

Dalteparin. In a single study evaluating therapeutic-dose dalteparin, comparison of the mean 4-hour anti-Xalevels in patients with a creatinine clearance of 40 mL/minor less versus normal renal function showed no differencebetween the groups (49).

Bleeding Events

Major bleeding events in patients with and withoutsevere renal insufficiency were reported in 12 studies in-volving 4971 patients. Ten studies involving 4741 patients

used enoxaparin (17–19, 24, 26–30, 39), and 2 studiesinvolving 230 patients used tinzaparin (36, 53). Use ofLMWH in patients with a creatinine clearance of 30 mL/min or less versus those with a creatinine clearance greaterthan 30 mL/min was associated with an increased risk formajor bleeding (5.0% vs. 2.4%; odds ratio, 2.25 [CI, 1.19to 4.27]; P � 0.013) (Figure 2). There was evidence ofstatistical heterogeneity (I2 � 50.4%; P � 0.028).

For the secondary analysis according to LMWH type,use of enoxaparin in patients with severe renal insufficiencyincreased the rate of major bleeding (6.0% vs. 2.4%; oddsratio, 2.59 [CI, 1.34 to 5.01]) (Figure 3). The data forbleeding events were not pooled in the 2 tinzaparin studiesbecause the odds ratios for bleeding could not be calculatedfor 1 study that had no bleeding events. We also could notpool these data for the prophylactic-dose studies becausethey were available for only 1 study (28). Among therapeu-tic-dose enoxaparin studies, major bleeding was statisticallysignificantly higher among patients with a creatinine clear-ance of 30 mL/min or less versus those with a creatinineclearance greater than 30 mL/min (8.3% vs. 2.4%; oddsratio, 3.88 [CI, 1.78 to 8.45]). When enoxaparin doseswere adjusted empirically according to creatinine clearanceor measured anti-Xa levels, the odds ratio for major bleed-ing was lower, although the CIs were broad (0.9% vs.1.9%; odds ratio, 0.58 [CI, 0.09 to 3.78]).

In the sensitivity analysis, all prospective studies re-ported consecutive patient enrollment and follow-up formajor bleeding was reported as “in hospital” for almost allstudies; therefore, we could not separate studies by qualityon the basis of these 2 characteristics. The sensitivity analy-sis was thus based on the use of bleeding definitions de-fined a priori. After the 2 studies that did not state defini-tions a priori (18, 26) were excluded, the odds ratio formajor bleeding in patients with severe renal insufficiencyreceiving LMWH was 2.17 (CI, 1.14 to 4.16).

Results obtained by using exact statistical methods for

Table 1—Continued


Study Designand Method ofEnrollment

FundingSource

Indication forLMWH


Patients,n

Mean Dose MajorBleeding,%‡

Bleeding Follow-up§:Mean Duration;Completeness

Siguret et al.,2000 (53)


NS AF, VTE, other 20–2930–3940–49�50

8967

175 IU/kg per d 0000

10 d; no loss tofollow-up

DalteparinShprecher et

al., 2005 (49)Prospective

cohort; CNS ACS, AF, VTE �40

�801111

100 IU/kg every 12 h Notassessed

–

* ACS � acute coronary syndrome; AF � atrial fibrillation; ESSENCE � Efficacy and Safety of Subcutaneous Enoxaparin in Non-Q-wave Coronary Events; IV �intravenous; LMWH � low-molecular-weight heparin; NA � not applicable; NC � nonconsecutive (not applicable to retrospective cohorts); NS � not specified; PK �pharmacokinetic; TIMI � Thrombolysis in Myocardial Infarction; VTE � venous thromboembolism.† Creatinine clearance calculated by Cockcroft–Gault equation except in Thorveska et al. (27) and Chow et al. (30) (Modification of Diet in Renal Disease equation),Sanderink et al. (38) (endogenous creatinine clearance), and Cadroy et al. (45) (Kampmann equation).‡ Bleeding assessed using TIMI criteria for bleeding from references 19, 29, and 31.§ Completeness of follow-up not applicable to retrospective studies and studies that assessed anti-Xa only.




Table 2. Measured Anti-Xa Levels and Relation to Renal Function*


Dose Timing of Anti-XaMeasurement


Anti-Xa Level and DosingSchedule, IU/mL

Conclusions

EnoxaparinTherapeutic dose

Bazinet et al.,2005 (17)

1 mg/kg every12 h or 1.5mg/kg per d

After 2nd or 3rddose, based on adaily dosingschedule; after4th or 5th dose,based on atwice-daily dosingschedule

10–30 1.18 (daily dosing);1.27 (twice-daily dosing)

Anti-Xa decreases with higher creatinineclearance (�0.003 IU/mL for each mL/min);twice-daily dosing can result insupratherapeutic anti-Xa levels in patientswith creatinine clearance �30 mL/min

31–50 1.21 (daily dosing);1.25 (twice daily dosing)

�50 1.10 (daily dosing);1.06 (twice-daily dosing)

Peng et al.,2004 (26)

1 mg/kg every12 h

After 3rd dose(mean, 5.2 doses)

�30 1.58 (SD, 0.54) Significantly higher peak anti-Xa levels arefound in patients with renal insufficiency;dose adjustment and/or laboratorymonitoring is recommended with creatinineclearance �30 mL/min

31–60 1.25 (SD, 0.38)

�60 0.93 (SD, 0.28)

Chow et al.,2003 (30)

1 mg/kg every12 h

After 3rd dose �30 1.34 Inverse correlation between creatinineclearance and peak anti-Xa levels; doseadjustments recommended for creatinineclearance �30 mL/min

�30 0.91

Becker et al.,2002 (37)

1 mg/kg or 1.25mg/kg every12 h

After 3rd or 5thdose, timing notspecified

�40 1.58 (3rd dose); 1.53(5th dose)

In patients with creatinine clearance �30mL/min compared with those with normalrenal function, 1) clearance reduced by22%; 2) higher trough and peak anti-Xaactivity; 3) dose adjustment necessary

40–80 1.41 (3rd dose);1.53 (5th dose)

�80 1.25 (3rd dose); 1.29(5th dose)

Adjusted doseKruse and Lee,

2004 (19)1 mg/kg, then

0.5 or 0.75mg/kg every12 h

After 3rd dose �30 0.65 (SD, 0.19) Adjusted dose produces therapeutic-rangeanticoagulation with bleeding incidencesimilar to normal renal function

New dose � current dose � goal anti-Xa/current anti-Xa

30–60 0.82 (SD, 0.18)

Collet et al.,2003 (31)‡

1 mg/kg, then65% of dose

After 3rd dose �30 0.85 (SD, 0.05) Adjusted dose provides adequate anti-Xa levelsand no excess of bleeding in patients withcreatinine clearance �30 mL/min�30 0.95 (SD, 0.02)


1 mg/kg, then65% of dose

Before cardiaccatheterization(mean, 5.4 h [SD,0.2] after 3.9 d[SD, 0.4])

�0 0.95 (SD, 0.07) Adjusted dose results in a safe level ofanticoagulation

31–60 0.95 (SD, 0.05)

�60 1.01 (SD, 0.05)

Prophylactic doseMahe et al.,

2002 (34)4000 IU/d After 1st or 2nd

dose�40 0.49 No correlation between anti-Xa activity and

creatinine clearance�40 0.52Sanderink et al.,

2002 (38)40 mg/d After 1st and 4th

dose�30 0.46 (day 1); 0.58 (day 4) In patients with creatinine clearance �30

mL/min compared with those with normalrenal function, 1) elimination decreased by35% and more evident with repeateddosing; 2) increased anti-Xa exposure causedby a slower rate of elimination, not bydifferences in absorption; 3) may need doseadjustment

31–50 0.45 (day 1); 0.50 (day 4)

51–80 0.49 (day 1); 0.56 (day 4)

�80 0.39 (day 1); 0.42 (day 4)

Cadroy et al.,1991 (45)

0.5 mg/kg(single dose)

After single dose �20 Mean, 0.35 (SD, 0.07) Decreased elimination in patients withcreatinine clearance �20 mL/min comparedwith healthy persons (higher Amax, longerTmax, increased area under thereceiver-operating curve, reduced clearance,and longer elimination of half-life)

�85 Mean, 0.29 (SD, 0.06)

TinzaparinPautas et al.,

2002 (36)175 IU/kg per d 5 h after the 1st

dose and onceweekly

20–34 0.69 (day 2); 0.87 (day 10);0.83 (day 20); 0.90 (day 28)

No correlation between anti-Xa activity andcreatinine clearance or age for each group

35–49 0.81 (day 2); 0.85 (day 10);0.82 (day 20); 0.84 (day 28)

50–64 0.81 (day 2); 0.85 (day 10);0.81 (day 20); 0.78 (day 28)

�65 0.83 (day 2); 0.84 (day 10);0.89 (day 20); 1.09 (day 28)

Continued on following page




stratified 2 � 2 tables to calculate a pooled risk differencewere consistent with those obtained by calculating the Petoodds ratios and therefore yielded similar conclusions (re-sults not presented).

Assessment of Publication Bias

A funnel plot of relative risk versus standard error in-cluded 9 of the 12 studies (17–19, 24, 27–29, 36, 39)because no bleeding events occurred in 3 studies. The fun-nel plot was asymmetrical, with an absence of studies in thebottom right-hand corner (data not shown).

DISCUSSION

Our results show that when standard therapeutic-doseLMWH (enoxaparin) is used in patients with severe renalinsufficiency (creatinine clearance �30 mL/min) com-pared with those without (creatinine clearance �30 mL/min), anti-Xa levels are higher and the risk for major bleed-ing events is increased 2- to 3-fold. Because almost allincluded studies used enoxaparin, these data suggest thatstandard, weight-adjusted, therapeutic-dose, unmonitoredenoxaparin should be avoided in patients with severe renalinsufficiency. There are insufficient data to make conclu-sions regarding tinzaparin, dalteparin, or other LMWHpreparations.

In patients with unstable angina and non–ST-segmentelevation myocardial infarction, suboptimal anticoagula-tion with LMWH (anti-Xa level �0.5 IU/mL) is associ-ated with an increased risk for recurrent ischemic eventsand is an independent predictor of mortality (21). Ourresults suggest that when the enoxaparin dose in patientswith acute coronary syndromes and severe renal insuffi-ciency is adjusted, either empirically or on the basis ofanti-Xa heparin levels, therapeutic anti-Xa levels can beachieved and excessive levels can be avoided. However, the

odds ratio for adjusted-dose enoxaparin was associated withwide CIs; therefore, we cannot definitely conclude thatempirically dose-adjusted enoxaparin is not associated withincreased risks for major bleeding events. This strategymerits further evaluation, both in patients with acute cor-onary syndromes and in those with venous thromboembo-lism who have severe renal insufficiency.

Analysis by study quality did not affect the results.Although the method of patient enrollment and durationof follow-up were similar in all studies, the sensitivity analy-sis, which included only studies using a priori bleedingdefinitions, resulted in a similar 2- to 3-fold increase inmajor bleeding when LMWH was used in patients withsevere renal insufficiency.

The data for tinzaparin in patients with severe renalinsufficiency are limited. There was no correlation betweencreatinine clearance and anti-Xa levels and no statisticallysignificant increase in anti-Xa levels after 10 days of ther-apeutic-dose tinzaparin in elderly patients with a mean cre-atinine clearance of 41 mL/min, but anti-Xa levels werenot compared between patients with and without renalinsufficiency (36, 53). The data for dalteparin also werelimited. Although there was no difference in the meananti-Xa levels in patients with and without renal insuffi-ciency, this was based on the mean of 1 measurement inonly a small number of patients (49).

We cannot make definitive conclusions concerning theappropriate threshold below which clinicians should beconcerned about the risk for LMWH accumulation. Pop-ulation pharmacokinetic analyses using anti-Xa levels (32,40) followed by simulations to determine a dosing strategyhave been published (18, 51). However, these strategieshave not been evaluated in patients with renal insuffi-ciency. The risk for excessive levels of anti-Xa is higherwhen therapeutic doses are used, but multiple prophylactic

Table 2—Continued


Dose Timing of Anti-XaMeasurement


Anti-Xa Level and DosingSchedule, IU/mL

Conclusions

Siguret et al.,2000 (53)

175 IU/kg per d 5 h after 2nd and10th dose

20–29 0.73 (SD, 0.16) (day 2);0.77 (SD, 0.19) (day 10)

No correlation between anti-Xa activity andcreatinine clearance and no accumulation ofanti-Xa activity30–39 0.57 (SD, 0.26) (day 2);

0.60 (SD, 0.21) (day 10)40–49 0.72 (SD, 0.22) (day 2);

0.75 (SD, 0.19) (day 10)�50 0.65 (SD, 0.14) (day 2);

0.71 (SD, 0.19) (day 10)

DalteparinSchprecher et al.,

2005 (45)100 IU/kg every

12 h3 to 5 h after 5th

or 6th dose�40 Mean, 0.47 (SD, 0.25) No difference in anti-Xa activity between

groups�80 Mean, 0.55 (SD, 0.20)

* Anti-Xa measurement at 4 hours after injection, unless indicated. Amax � maximum low-molecular-weight heparin activity; Tmax � time of maximum observed activity.† Creatinine clearance calculated by Cockcroft–Gault equation except in Chow et al. (30) (Modification of Diet in Renal Disease equation), Sanderink et al. (38)(endogenous creatinine clearance), and Schprecher et al. (45) (Kampmann equation).‡ Doses calculated as follows: creatinine clearance �30 mL/min received 65% of recommended dose, anti-Xa measured 4 hours after the 3rd injection, and enoxaparin doseadjusted aiming for anti-Xa levels of 0.5 to 1.0 IU/mL.




doses are still associated with LMWH accumulation, andlaboratory monitoring or dose reduction may also be nec-essary in patients with severe renal insufficiency. This isparticularly relevant for patients who require long-termprophylactic-dose LMWH therapy.

There were considerable differences among the studiesincluded in our review, including age, concomitant anti-platelet therapy (in the studies evaluating enoxaparin foracute coronary syndromes), number of doses of LMWHadministered, and frequency of bleeding outcome assess-ment, all of which affect the risk for bleeding. Further-more, differences in the methods used to calculate creati-nine clearance may have affected the proportion of patientsdefined as having severe renal insufficiency. Finally, al-though all studies used chromogenic anti-Xa assays, differ-ent assays and instruments can produce different anti-Xaresults (55).

In the Global Registry of Acute Coronary Events(GRACE) study (50), major bleeding events duringLMWH therapy occurred in 5.90% of patients with severerenal insufficiency (creatinine clearance �30 mL/min), in2.03% of patients with moderate renal insufficiency (cre-atinine clearance 30 to 60 mL/min), and in 1.24% of pa-

tients with normal renal function (creatinine clearance�60 mL/min). This study did not meet our inclusion cri-teria because various LMWH preparations and doses wereused. However, a pooled analysis including these data didnot alter our results or conclusions: Major bleeding oc-curred in 5.4% of patients with a creatinine clearance of 30mL/min or less compared with 2.0% of patients with acreatinine clearance greater than 30 mL/min (odds ratio,3.85 [CI, 2.29 to 6.48]; P � 0.001).

Our study has several potential limitations. First, al-though most studies enrolled consecutive patients, the po-tential for confounding by indication cannot be excluded.For example, patients at lower risk for bleeding events mayhave been preferentially included, which may lead to anunderestimate of the risk for these events associated withLMWH in patients with severe renal insufficiency (9). Tominimize potential biases, we excluded case reports andselected studies with 10 or more patients. Second, rela-tively few patients with severe renal insufficiency were in-cluded in our analyses, and the number of bleeding eventswas modest. Third, despite careful electronic and manualsearches and contact with content experts to identify allavailable data, the funnel plot of study size versus treat-

Figure 2. Peto odds ratio (OR) of major bleeding events in patients with severe renal insufficiency (creatinine clearance <30mL/min) compared with patients without renal insufficiency (creatinine clearance >30 mL/min).




ment effect was asymmetrical, which suggests publicationbias attributable to underreporting of small studies show-ing an increased risk for major bleeding events. However,this probably underestimates rather than overestimates ma-

jor bleeding events when LMWH is used in patients withsevere renal insufficiency. Fourth, although the point esti-mate for major bleeding events when adjusted-dose enox-aparin is used in patients with severe renal insufficiency

Figure 3. Peto odds ratio (OR) of major bleeding events with enoxaparin in patients with severe renal insufficiency (creatinineclearance <30 mL/min) compared with patients without renal insufficiency (creatinine clearance >30 mL/min).




suggests that this strategy is not associated with increasedrisks, there may still be a clinically important increase inrisk because lack of evidence of a difference is not the sameas evidence of no difference. Furthermore, the comparisonbetween standard-dose and adjusted-dose enoxaparin wasnot randomized. Finally, although our review included 3LMWHs, most of the available data were for enoxaparin;data for tinzaparin and dalteparin are limited, precludingdefinitive conclusions regarding the use of these and otherLMWHs in patients with severe renal insufficiency.

Our findings have important clinical implications.Enoxaparin is widely used in patients with acute coronarysyndromes and acute venous thromboembolism. Approxi-mately 50% of these patients have impaired renal function,and a substantial minority have severe renal insufficiency(50, 56, 57). Our results suggest that renal function shouldbe measured in all patients requiring therapeutic-doseLMWH and that the dose should be reduced or the an-ti-Xa level should be monitored in patients with a creati-nine clearance of 30 mL/min or less. Prophylactic-doseLMWH should also be used with caution in patients withsevere renal insufficiency, and empirical dose reductionmay be prudent.

Future studies should examine the pharmacokineticprofiles of different LMWHs in patients with varying de-grees of renal insufficiency to better assess the risk for ac-cumulation and bleeding events during extended treatmentwith prophylactic and therapeutic doses of LMWH. Ran-domized trials are needed to compare different doses ofLMWH in patients with varying degrees of renal insuffi-ciency to determine the optimal anticoagulant strategy thatminimizes the risk for bleeding complications while main-taining antithrombotic efficacy.

In conclusion, standard, weight-adjusted LMWH(enoxaparin) is associated with a 2- to 3-fold increased riskfor major bleeding events in patients with severe renal in-sufficiency (creatinine clearance �30 mL/min) versus pa-tients without renal insufficiency. Adjusted doses ofLMWH may reduce the risk for bleeding events in thesepatients but needs to be further evaluated.

From McMaster University and Hamilton General Hospital, Hamilton,Ontario, Canada, and University of Insubria, Varese, Italy.

Acknowledgments: The authors thank Dr. Qilong Li for providingadditional statistical analyses and the authors of the primary studies, Drs.L. Lalonde, I. Mahe, V. Siguret, and N. Thorevska, who kindly providedinformation included in the meta-analysis.

Grant Support: Dr. Crowther is a Career Investigator of the Heart andStroke Foundation of Canada; Dr. Eikelboom holds a Tier II CanadaResearch Chair in Cardiovascular Medicine from the Canadian Institutesfor Health Research; and Dr. Lim is the recipient of a Graduate Schol-arship from the Canadian Institutes of Health Research.

Potential Financial Conflicts of Interest: Consultancies: M.A. Crowther(AstraZeneca, Pfizer/Pharmacia, Sanofi-Aventis, GlaxoSmithKline, LaJolla Pharmaceutical Corporation, Leo Laboratories, Sandoz); Honoraria:

J.W. Eikelboom (Pharmacia Corp., Sanofi, Aventis), M.A. Crowther(Pfizer/Pharmacia, Sanofi-Aventis, GlaxoSmithKline, Leo Laboratories,Calea, Novo Nordisk); Grants received: W. Lim (Sanofi-Aventis), J.W.Eikelboom (Sanofi, Aventis), M.A. Crowther (Pfizer/Pharmacia, Sanofi-Aventis, Leo Laboratories).

Corresponding Author: Wendy Lim, MD, BSc, St. Joseph’s Hospital,50 Charlton Avenue East, Room L208, Hamilton, Ontario L8N 4A6,Canada; e-mail, [email protected].

Current author addresses are available at www.annals.org.

References1. Quinlan DJ, McQuillan A, Eikelboom JW. Low-molecular-weight heparincompared with intravenous unfractionated heparin for treatment of pulmonaryembolism: a meta-analysis of randomized, controlled trials. Ann Intern Med.2004;140:175-83. [PMID: 14757615]2. Gould MK, Dembitzer AD, Doyle RL, Hastie TJ, Garber AM. Low-molec-ular-weight heparins compared with unfractionated heparin for treatment ofacute deep venous thrombosis. A meta-analysis of randomized, controlled trials.Ann Intern Med. 1999;130:800-9. [PMID: 10366369]3. Boneu B, Caranobe C, Sie P. Pharmacokinetics of heparin and low molecularweight heparin. Baillieres Clin Haematol. 1990;3:531-44. [PMID: 2176903]4. Hirsh J, Raschke R. Heparin and low-molecular-weight heparin: the SeventhACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest. 2004;126:188S-203S. [PMID: 15383472]5. Busby LT, Weyman A, Rodgers GM. Excessive anticoagulation in patientswith mild renal insufficiency receiving long-term therapeutic enoxaparin. Am JHematol. 2001;67:54-6. [PMID: 11279659]6. Farooq V, Hegarty J, Chandrasekar T, Lamerton EH, Mitra S, HoughtonJB, et al. Serious adverse incidents with the usage of low molecular weight hepa-rins in patients with chronic kidney disease. Am J Kidney Dis. 2004;43:531-7.[PMID: 14981611]7. Laposata M, Green D, Van Cott EM, Barrowcliffe TW, Goodnight SH,Sosolik RC. College of American Pathologists Conference XXXI on laboratorymonitoring of anticoagulant therapy: the clinical use and laboratory monitoringof low-molecular-weight heparin, danaparoid, hirudin and related compounds,and argatroban. Arch Pathol Lab Med. 1998;122:799-807. [PMID: 9740137]8. Nagge J, Crowther M, Hirsh J. Is impaired renal function a contraindicationto the use of low-molecular-weight heparin? Arch Intern Med. 2002;162:2605-9.[PMID: 12456233]9. Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, etal. Meta-analysis of observational studies in epidemiology: a proposal for report-ing. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group.JAMA. 2000;283:2008-12. [PMID: 10789670]10. Cockcroft DW, Gault MH. Prediction of creatinine clearance from serumcreatinine. Nephron. 1976;16:31-41. [PMID: 1244564]11. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A moreaccurate method to estimate glomerular filtration rate from serum creatinine: anew prediction equation. Modification of Diet in Renal Disease Study Group.Ann Intern Med. 1999;130:461-70. [PMID: 10075613]12. Harenberg J. Is laboratory monitoring of low-molecular-weight heparin ther-apy necessary? Yes. J Thromb Haemost. 2004;2:547-50. [PMID: 15102006]13. Frydman A. Low-molecular-weight heparins: an overview of their pharma-codynamics, pharmacokinetics and metabolism in humans. Haemostasis.1996;26 Suppl 2:24-38. [PMID: 8707165]14. Cornelli U, Fareed J. Human pharmacokinetics of low molecular weightheparins. Semin Thromb Hemost. 1999;25 Suppl 3:57-61. [PMID: 10549717]15. Maclure M, Willett WC. Misinterpretation and misuse of the kappa statistic.Am J Epidemiol. 1987;126:161-9. [PMID: 3300279]16. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysisdetected by a simple, graphical test. BMJ. 1997;315:629-34. [PMID: 9310563]17. Bazinet A, Almanric K, Brunet C, Turcotte I, Martineau J, Caron S, et al.Dosage of enoxaparin among obese and renal impairment patients. Thromb Res.2005;116:41-50. [PMID: 15850607]18. Green B, Greenwood M, Saltissi D, Westhuyzen J, Kluver L, Rowell J, etal. Dosing strategy for enoxaparin in patients with renal impairment presentingwith acute coronary syndromes. Br J Clin Pharmacol. 2005;59:281-90. [PMID:




15752373]19. Kruse MW, Lee JJ. Retrospective evaluation of a pharmacokinetic programfor adjusting enoxaparin in renal impairment. Am Heart J. 2004;148:582-9.[PMID: 15459586]20. Ma JM, Jackevicius CA, Yeo E. Anti-Xa monitoring of enoxaparin for acutecoronary syndromes in patients with renal disease. Ann Pharmacother. 2004;38:1576-81. [PMID: 15328396]21. Montalescot G, Collet JP, Tanguy ML, Ankri A, Payot L, Dumaine R, etal. Anti-Xa activity relates to survival and efficacy in unselected acute coronarysyndrome patients treated with enoxaparin. Circulation. 2004;110:392-98.[PMID 15249498]22. Hulot JS, Vantelon C, Urien S, Bouzamondo A, Mahe I, Ankri A, et al.Effect of renal function on the pharmacokinetics of enoxaparin and consequenceson dose adjustment. Ther Drug Monit. 2004;26:305-10. [PMID: 15167633]23. Brophy DF, Martin EJ, Gehr TW, Carr ME Jr. Enhanced anticoagulantactivity of enoxaparin in patients with ESRD as measured by thrombin genera-tion time. Am J Kidney Dis. 2004;44:270-7. [PMID: 15264185]24. Macie C, Forbes L, Foster GA, Douketis JD. Dosing practices and riskfactors for bleeding in patients receiving enoxaparin for the treatment of an acutecoronary syndrome. Chest. 2004;125:1616-21. [PMID: 15136367]25. Desjardins L, Bara L, Boutitie F, Samama MM, Cohen AT, Combe S, et al.Correlation of plasma coagulation parameters with thromboprophylaxis, patientcharacteristics, and outcome in the MEDENOX study. Arch Pathol Lab Med.2004;128:519-26. [PMID: 15086284]26. Peng YG, Eikelboom JW, Tenni P, McQuillan A, Thom J. Renal function,peak anti-xa levels and enoxaparin dosing. J Pharm Practice Res. 2004;34:14-17.27. Thorevska N, Amoateng-Adjepong Y, Sabahi R, Schiopescu I, Salloum A,Muralidharan V, et al. Anticoagulation in hospitalized patients with renal insuf-ficiency: a comparison of bleeding rates with unfractionated heparin vs enoxapa-rin. Chest. 2004;125:856-63. [PMID: 15006942]28. Khazan M, Scheuering S, Adamson R, Mathis AS. Prescribing patterns andoutcomes of enoxaparin for anticoagulation of atrial fibrillation. Pharmacother-apy. 2003;23:651-8. [PMID: 12741440]29. Spinler SA, Inverso SM, Cohen M, Goodman SG, Stringer KA, AntmanEM, et al. Safety and efficacy of unfractionated heparin versus enoxaparin inpatients who are obese and patients with severe renal impairment: analysis fromthe ESSENCE and TIMI 11B studies. Am Heart J. 2003;146:33-41. [PMID:12851605]30. Chow SL, Zammit K, West K, Dannenhoffer M, Lopez-Candales A. Cor-relation of antifactor Xa concentrations with renal function in patients on enox-aparin. J Clin Pharmacol. 2003;43:586-90. [PMID: 12817521]31. Collet JP, Montalescot G, Fine E, Golmard JL, Dalby M, Choussat R, et al.Enoxaparin in unstable angina patients who would have been excluded fromrandomized pivotal trials. J Am Coll Cardiol. 2003;41:8-14. [PMID: 12570937]32. Bruno R, Baille P, Retout S, Vivier N, Veyrat-Follet C, Sanderink GJ, et al.Population pharmacokinetics and pharmacodynamics of enoxaparin in unstableangina and non-ST-segment elevation myocardial infarction. Br J Clin Pharma-col. 2003;56:407-14. [PMID: 12968985]33. Cestac P, Bagheri H, Lapeyre-Mestre M, Sie P, Fouladi A, Maupas E, et al.Utilisation and safety of low molecular weight heparins: prospective observationalstudy in medical inpatients. Drug Saf. 2003;26:197-207. [PMID: 12580648]34. Mahe I, Drouet L, Chassany O, Grenard AS, Caulin C, Bergmann JF. Lowmolecular weight heparin for the prevention of deep venous thrombosis: a suit-able monitoring in elderly patients? Pathophysiol Haemost Thromb. 2002;32:134-6. [PMID: 12372928]35. Choussat R, Montalescot G, Collet JP, Vicaut E, Ankri A, Gallois V, et al.A unique, low dose of intravenous enoxaparin in elective percutaneous coronaryintervention. J Am Coll Cardiol. 2002;40:1943-50. [PMID: 12475453]36. Pautas E, Gouin I, Bellot O, Andreux JP, Siguret V. Safety profile oftinzaparin administered once daily at a standard curative dose in two hundredvery elderly patients. Drug Saf. 2002;25:725-33. [PMID: 12167068]37. Becker RC, Spencer FA, Gibson M, Rush JE, Sanderink G, Murphy SA, etal. Influence of patient characteristics and renal function on factor Xa inhibitionpharmacokinetics and pharmacodynamics after enoxaparin administration innon-ST-segment elevation acute coronary syndromes. Am Heart J. 2002;143:753-9. [PMID: 12040334]38. Sanderink GJ, Guimart CG, Ozoux ML, Jariwala NU, Shukla UA, Bout-ouyrie BX. Pharmacokinetics and pharmacodynamics of the prophylactic dose of

enoxaparin once daily over 4 days in patients with renal impairment. ThrombRes. 2002;105:225-31. [PMID: 11927128]39. Collet JP, Montalescot G, Choussat R, Lison L, Ankri A. Enoxaparin inunstable angina patients with renal failure [Letter]. Int J Cardiol. 2001;80:81-2.[PMID: 11575265]40. Barrett JS, Gibiansky E, Hull RD, Planes A, Pentikis H, Hainer JW, et al.Population pharmacodynamics in patients receiving tinzaparin for the preventionand treatment of deep vein thrombosis. Int J Clin Pharmacol Ther. 2001;39:431-46. [PMID: 11680668]41. Brophy DF, Wazny LD, Gehr TW, Comstock TJ, Venitz J. The pharma-cokinetics of subcutaneous enoxaparin in end-stage renal disease. Pharmacother-apy. 2001;21:169-74. [PMID: 11213853]42. Pautas E, Siguret V, d’Urso M, Laurent M, Gaussem P, Fevrier M, et al.[Monitoring of tinzaparin in a ten day treatment dose in elderly patients]. RevMed Interne. 2001;22:120-6. [PMID: 11234669]43. Gerlach AT, Pickworth KK, Seth SK, Tanna SB, Barnes JF. Enoxaparinand bleeding complications: a review in patients with and without renal insuffi-ciency. Pharmacotherapy. 2000;20:771-5. [PMID: 10907967]44. Mismetti P, Laporte-Simitsidis S, Navarro C, Sie P, d’Azemar P, NecciariJ, et al. Aging and venous thromboembolism influence the pharmacodynamics ofthe anti-factor Xa and anti-thrombin activities of a low molecular weight heparin(nadroparin). Thromb Haemost. 1998;79:1162-5. [PMID: 9657442]45. Cadroy Y, Pourrat J, Baladre MF, Saivin S, Houin G, Montastruc JL, et al.Delayed elimination of enoxaparin in patients with chronic renal insufficiency.Thromb Res. 1991;63:385-90. [PMID: 1659748]46. Goudable C, Saivin S, Houin G, Sie P, Boneu B, Tonthat H, et al. Phar-macokinetics of a low molecular weight heparin (Fraxiparine) in various stages ofchronic renal failure. Nephron. 1991;59:543-5. [PMID: 1662782]47. Hory B, Claudet MH, Magnette J, Bechtel P, Bayrou B. Pharmacokineticof a very low molecular weight heparin in chronic renal failure. Thromb Res.1991;63:311-7. [PMID: 1659746]48. Follea G, Laville M, Pozet N, Dechavanne M. Pharmacokinetic studies ofstandard heparin and low molecular weight heparin in patients with chronic renalfailure. Haemostasis. 1986;16:147-51. [PMID: 3710292]49. Shprecher AR, Cheng-Lai A, Madsen EM, Cohen HW, Sinnett MJ, WongST, et al. Peak antifactor xa activity produced by dalteparin treatment in patientswith renal impairment compared with controls. Pharmacotherapy. 2005;25:817-22. [PMID: 15927900]50. Collet JP, Montalescot G, Agnelli G, Van de Werf F, Gurfinkel EP, Lopez-Sendon J, et al. Non-ST-segment elevation acute coronary syndrome in patientswith renal dysfunction: benefit of low-molecular-weight heparin alone or withglycoprotein IIb/IIIa inhibitors on outcomes. The Global Registry of Acute Cor-onary Events. Eur Heart J. 2005;26:2285-93. [PMID: 15932908]51. Hulot JS, Montalescot G, Lechat P, Collet JP, Ankri A, Urien S. Dosingstrategy in patients with renal failure receiving enoxaparin for the treatment ofnon-ST-segment elevation acute coronary syndrome. Clin Pharmacol Ther.2005;77:542-52. [PMID: 15961985]52. Kane-Gill SL, Feng Y, Bobek MB, Bies RR, Pruchnicki MC, Dasta JF.Administration of enoxaparin by continuous infusion in a naturalistic setting:analysis of renal function and safety. J Clin Pharm Ther. 2005;30:207-13.[PMID: 15896237]53. Siguret V, Pautas E, Fevrier M, Wipff C, Durand-Gasselin B, Laurent M,et al. Elderly patients treated with tinzaparin (Innohep) administered once daily(175 anti-Xa IU/kg): anti-Xa and anti-IIa activities over 10 days. Thromb Hae-most. 2000;84:800-4. [PMID: 11127859]54. Aventis Pharma Inc. Lovenox product monograph (enoxaparin sodium).2004.55. Kovacs MJ, Keeney M. Inter-assay and instrument variability of anti-Xa—results [Letter]. Thromb Haemost. 2000;84:138. [PMID: 10928485]56. Gibson CM, Dumaine RL, Gelfand EV, Murphy SA, Morrow DA, WiviottSD, et al. Association of glomerular filtration rate on presentation with subse-quent mortality in non-ST-segment elevation acute coronary syndrome; observa-tions in 13,307 patients in five TIMI trials. Eur Heart J. 2004;25:1998-2005.[PMID: 15541835]57. Buller HR, Davidson BL, Decousus H, Gallus A, Gent M, Piovella F, et al.Subcutaneous fondaparinux versus intravenous unfractionated heparin in the ini-tial treatment of pulmonary embolism. N Engl J Med. 2003;349:1695-702.[PMID: 14585937]




Current Author Addresses: Drs. Lim and Crowther: St. Joseph’s Hos-pital, 50 Charlton Avenue East, Room L208, Hamilton, Ontario L8N4A6, Canada.Dr. Dentali: Department of Medicine, University of Insubria, VialeBorri 57, 21100 Varese, Italy.Dr. Eikelboom: Hamilton Health Sciences, Hamilton General Hospital,237 Barton Street East, Hamilton, Ontario L8L 2X2, Canada.

APPENDIX: MEDLINE SEARCH STRATEGY

Database: Ovid MEDLINE(R) � 1966 to November Week3 2005 �

Search Strategy:1. low molecular weight heparin.mp. or exp Heparin, Low-

Molecular-Weight/ (6850)2. dalteparin$.mp. or exp Tedelparin/ (653)3. (tedelparin$ or fragmin$ or kabi2165 or kabi 2165 or

k2165 or k 2165 or fr860 or fr 860).mp. [mp � title, originaltitle, abstract, name of substance word, subject heading word](729)

4. enoxaparin$.mp. or exp ENOXAPARIN/ (1604)5. (lovenox or clexane or klexane or pk10169 or pk 10169

or emt996 or emt 996 or emt967 or emt 967).mp. [mp�title,original title, abstract, name of substance word, subject headingword] (155)

6. nadroparin$.mp. or exp NADROPARIN/ (360)7. (fraxiparin$ or seleparin$ or tedegliparin$ or cy216 or cy

216).mp. [mp�title, original title, abstract, name of substanceword, subject heading word] (268)

8. tinzaparin$.mp. (160)9. (innohep or logiparin$).mp. [mp � title, original title,

abstract, name of substance word, subject heading word] (50)10. antixarin.mp. (1)11. (ardeparin$ or normiflo or rd11885 or rd 11885).mp.

[mp�title, original title, abstract, name of substance word, sub-ject heading word] (33)

12. (bemiparin$ or hibor or ivor or zibor or badyket).mp.[mp�title, original title, abstract, name of substance word, sub-ject heading word] (128)

13. (certoparin$ or alpharin$ or sandoparin$ or troparin$ orembolex or monoembolex).mp. [mp � title, original title, ab-stract, name of substance word, subject heading word] (69)

14. (parnaparin$ or fluxum or op2123 or op 2123 or min-idalton or alphaLMWH or alpha LMWH).mp. [mp�title, orig-inal title, abstract, name of substance word, subject headingword] (27)

15. (reviparin$ or lu473111 or lu 473111 or clivarin$).mp.[mp�title, original title, abstract, name of substance word, sub-ject heading word] (109)

16. or/1-15 (7229)17. kidney failure.mp. or exp Kidney Failure/ (78541)18. (renal failure or renal insufficiency or kidney insuffi-

ciency or renal function or kidney function or renal$ or kid-ney$).mp. [mp�title, original title, abstract, name of substanceword, subject heading word] (568417)

19. glomerular filtration rate.mp. or exp Glomerular Filtra-tion Rate/ (25533)

20. kidney diseases.mp. or exp Kidney Diseases/ (278443)21. or/17-20 (615811)22. 16 and 21 (541)23. limit 22 to humans (482)24. limit 23 to (case reports or clinical trial or controlled

clinical trial or multicenter study or randomized controlled trial)(219)

25. from 24 keep 1-219 (219)

Annals of Internal Medicine

W-158 2 May 2006 Annals of Internal Medicine Volume 144 • Number 9 www.annals.org


meta-analysis low-molecular-weight heparin and bleeding in patients.pdf

Documents

reportedcreatinine clearance

prophylactic doses of

lmwhtreated patients

levels of antixa heparin

major bleeding inpatients

data sources

data extraction

adjusted doses