ARTICLE

Evaluation of the need t

o discontinue antiplateletand anticoagulant medications before cataract

surgeryHiroshi Kobayashi, MD, PhD

Q 2010 A

Published

SCRS an

by Elsev

PURPOSE: To assess the risk for intraoperative and postoperative bleeding associated withantiplatelet and/or anticoagulant treatment in patients having uneventful phacoemulsification.

SETTING: Kokura Memorial Hospital, Kitakyusyu, Japan.

METHODS: In a nonrandomized case series, consecutive patients had phacoemulsification andintraocular lens implantation under sub-Tenon anesthesia. All patients were on warfarin,acetylsalicylic acid (aspirin) therapy, or both. Patients discontinued therapy 1 week beforesurgery (discontinuation group) or continued the therapeutic regimen until the time of surgery(maintenance group).

RESULTS: The discontinuation group comprised 182 patients and the maintenance group, 173patients. There was no significant difference between the 2 groups in the mean prothrombintime–international normalized ratio in patients taking warfarin (P Z .6). Although there was nosignificant intraoperative bleeding in any case, 47 eyes (16.5%) in the maintenance group and31 eyes (10.8%) in the discontinuation group had a subconjunctival hemorrhage postoperatively(P Z .0309). Minor postoperative ocular bleeding occurred in 11 eyes (4.0%) in the maintenancegroup and 7 eyes (2.5%) in the discontinuation group (P Z .4). During the 1-month postoperativeperiod, the mean change in corrected distance visual acuity was �0.462 logMAR G 0.331 (SD) inthe maintenance group and �0.434 G 0.318 logMAR in the discontinuation group (P Z .3).

CONCLUSIONS: Patients taking warfarin, aspirin, or both up to the time of phacoemulsification hada significantly higher incidence of subconjunctival hemorrhage than those who discontinuedtherapy. There was no significant difference between the 2 groups in the incidence of intraoperativeand postoperative complications or in visual improvement.

Financial Disclosure: The author has no financial or proprietary interest in any material or methodmentioned.

J Cataract Refract Surg 2010; 36:1115–1119 Q 2010 ASCRS and ESCRS

Patients receiving long-term antiplatelet and anticoag-ulant medications pose a clinical challenge when theneed for intraocular surgery arises, including cataractextraction. If antiplatelet and anticoagulation thera-pies are continued, there is a risk for serious bleedingcomplications; however, if these medications arediscontinued, there is a risk for thromboemboliccomplications.1–4 At present, there is little consensuson the appropriate perioperative treatment for pa-tients receiving long-term acetylsalicylic acid (aspirin)and warfarin therapy.

This study compared the incidence of hemorrhagicand nonhemorrhagic complications and the visualcourse after phacoemulsification and intraocular lens

d ESCRS

ier Inc.

(IOL) implantation in patients continuing antiplateletand/or anticoagulant medications and those discon-tinuing them.

PATIENTS AND METHODS

This study comprised consecutive patients taking warfarin,aspirin, or both for 6 months or longer who had cataractextraction between April 2006 and March 2008. Before April2007, all patients discontinued the drugs before surgery(discontinuation group). After April 2007, all patients main-tained the treatment at the time of surgery (maintenancegroup). Patients in the discontinuation group stoppedtherapy 1 week before surgery and resumed the therapy 2days postoperatively. A human subjects committeeapproved the study protocol and consent forms.

0886-3350/$dsee front matter 1115doi:10.1016/j.jcrs.2010.01.017

1116 ANTIPLATELET AND ANTICOAGULANT MEDICATIONS BEFORE CATARACT SURGERY

Surgical Technique

Phacoemulsification with IOL implantation wasperformed through a superior sclerocorneal incision afterscleral cauterization. Sub-Tenon anesthesia of lidocaine2.0% was administered using a Fukasaku blunt sub-Tenoncannula (Handaya Co., Ltd.). In all cases, a standardphacoemulsification technique was used and a 3-piecehydrophobic acrylic IOL (AcrySof MA30BM, Alcon, Inc.)implanted. Sodium hyaluronate 1.0% (Healon) was used asthe ophthalmic viscosurgical device.

Postoperatively, patients in both groups were giventopical levofloxacin, dexamethasone, and diclofenac 3 timesa day for 2 weeks, after which the drugs were tapered over 2months.

Evaluation of Outcomes

Table 1. Patient demographics.

Group

Parameter Maintenance Discontinuation P Value

Patients, n 173 182 d

Eyes, n 273 285 d

Age (y) .1005Mean G SD 73.3 G 8.6 74.7 G 7.4

Before enrollment, patients had a complete evaluationincluding an ocular and systemic history, slitlampbiomicroscopy, visual acuity, physical examination, dilatedfundoscopy, intraocular pressure (IOP) by Goldmannapplanation tonometry, and corrected distance visual acuity(CDVA) measurement. The CDVA was converted tologMAR values, which were used in all statistical analyses.The slitlamp findings were used to clinically grade thepreoperative nuclear sclerosis based on themethod of Emeryand Little.5 After pupil dilation, ophthalmoscopic andslitlamp biomicroscopic examinations were performed todetermine the presence of retinal, vitreous, and choroidal/suprachoroidal hemorrhages. After surgery, the patients’progress was reviewed at 1 and 3 days, 1 week, and 1month.

Intraoperatively and at all postoperative visits, patientswere assessed for the presence of general complications,including cerebral events (eg, transient ischemic attack,cerebral infarction, and hemorrhage) and cardiovascularevents (eg, deep vein thrombosis, myocardial ischemia,myocardial infarction). Hemorrhagic complications weredefined as a subconjunctival hemorrhage and hyphemaor as a retinal, vitreous, and choroidal/suprachoroidalhemorrhage. Hypotony was defined as an IOP lower than4 mm Hg after surgery. An IOP spike was defined as anIOP that was 3 mm Hg or higher on the first postoperativeday than preoperatively. The criteria of Teekhasaenee andRitch6 were used to define a shallow anterior chamber.

Range 46–88 48–90Sex, n .7599Female 95 97

Statistical Analysis Male 78 85

CDVADecimal 0.314 0.346 d

Mean 0.503 G 0.298 0.461 G 0.288 .0896

The Student t test was used to evaluate continuousvariables and the paired t test to evaluate the differences inIOP between the follow-up intervals. All t tests were 2-tailed.Categorical variables were evaluated using the chi-square

Submitted: June 24, 2009.Final revision submitted: December 29, 2009.Accepted: January 2, 2010.

From the Department of Ophthalmology, Kanmon Medical Center,Shimonoseki, Japan.

Corresponding author: Hiroshi Kobayashi, MD, PhD, Department ofOphthalmology, Kanmon Medical Center, 1-1 Chofu-Sotoura-cho,Shimonoseki 752-8510, Japan. E-mail: kobi@earth.ocn.ne.jp.

J CATARACT REFRACT SURG

test, Fisher exact test, or Spearman rank correlation. A Pvalue less than 0.05 was considered statistically significant.

Baseline age, sex, CDVA, and IOP values were used topair the 2 study groups. When correlations between pairedobservations were noted, the F test was used to determinethe variances between the groups.

RESULTS

Baseline

During the study period, 558 consecutive cataractextractions were performed in 355 patients. Table 1shows the patients’ demographics by group. Therewas no statistically significant difference in age, sex,or CDVA between the maintenance group and thediscontinuation group.

Complications

Table 2 shows the incidence of hemorrhagic and non-hemorrhagic complications. There were no systemiccomplications. The most frequent hemorrhagic compli-cation was a subconjunctival hemorrhage larger than1 quadrant; significantly more eyes in the maintenancegroup than in the discontinuation group developed thiscomplication (P Z .0309). In both groups, the hemor-rhageswere self-limiting and no eye required postoper-ative intervention.Minor postoperative ocular bleedingoccurred in 11 eyes (4.0%) in the maintenancegroup and 7 eyes (2.5%) in the discontinuation group

logMAR G SDNuclear sclerosis .0698Mean G SD 2.2 G 0.6 2.3 G 0.7Range 0–5 0–5

Administration, n (%) .7467Warfarin only 39 (63) 36 (59)Aspirin only 124 (198) 137 (213)Both 10 (15) 9 (15)

PT–INR 1.94 G 0.77 1.87 G 0.62 .6304

CDVA Z corrected distance visual acuity; PT–INR Z prothrombin time–international normalized ratio

- VOL 36, JULY 2010

Table 2. Incidence of hemorrhagic and nonhemorrhagic complications by group.

Eyes, n (%)

Complication Maintenance Discontinuation P Value

HemorrhagicSubconjunctival hemorrhage 47 (16.1) 31 (10.8) .0309Microscopic hyphema 1 d postop 8 (2.9) 5 (1.8) .3572Dot retinal hemorrhage 3 (1.1) 2 (0.7) .6187

NonhemorrhagicIntraoperative

Early perforation 2 (0.7) 3 (1.1) .6884CCC tear 10 (3.7) 8 (2.8) .5673Posterior capsule rupture 4 (1.5) 3 (1.1) .6616Vitreous loss 3 (1.1) 3 (1.1) .9578

Early postoperativeIOP spike 6 (2.2) 5 (1.8) .7064Corneal edema 11 (5.9) 13 (4.6) .7568

CCC Z continuous curvilinear capsulorhexis; IOP Z intraocular pressure

1117ANTIPLATELET AND ANTICOAGULANT MEDICATIONS BEFORE CATARACT SURGERY

(P Z .4). Within 1 week of surgery, all bleeding hadstopped without affecting visual acuity. There wereno cases of hyphema (O1.0 mm), vitreous hemorrhage,or choroidal/suprachoroidal hemorrhage in eithergroup.

The most frequent nonhemorrhagic intraoperativecomplication was a continuous curvilinear capsulo-rhexis tear. The most frequent nonhemorrhagiccomplication in the early postoperative period wascorneal edema. There was no significant differencebetween the maintenance group and the discontinua-tion group in the incidence of any nonhemorrhagicintraoperative or postoperative complication. Allnonhemorrhagic complications were self-limitingand did not require surgical intervention. There wereno cases of dropped nucleus intraoperatively or ofhypotony, shallow/flat anterior chamber, distortedpupil, IOL dislocation, vitreous herniation, retinaldetachment, or endophthalmitis postoperatively.

Visual Acuity

The mean decimal CDVA was 0.314 in the mainte-nance group and 0.346 in the discontinuation grouppreoperatively and 0.929 and 0.940, respectively, 1month postoperatively. The mean change from thepreoperative CDVA was �0.465 G 0.328 logMAR inthe maintenance group and �0.462 G 0.331 logMARin the discontinuation group 1 week postoperativelyand �0.431 G 0.315 logMAR and �0.434 G 0.318logMAR, respectively, 1 month postoperatively. Thedifferences between the 2 groups were not statisticallysignificant (P Z .2 and P Z .3, respectively).

J CATARACT REFRACT SURG

DISCUSSION

In this study, patients who continued taking warfarin,aspirin, or both had no increase in potentialsight-threatening complications associated withphacoemulsification and IOL implantation comparedwith patients who discontinued treatment 1 weekbefore surgery. Although the incidence of subconjunc-tival hemorrhagewas significantly greater in the main-tenance group than in the discontinuation group, thehemorrhages were self-limiting in both groups.Although patientswho continued therapy had a higherincidence of hyphema and retinal hemorrhage thanthose who discontinued therapy, the difference wasnot significant.

The results in several studies7–12 show thatmaintaining an antiplatelet and/or anticoagulantregimen during the perioperative period does notsignificantly increase the incidence of potentially sight-threatening complications associated with phacoemul-sification. However, patients who continue therapyhave a significantly higher incidence of subconjunctivalhemorrhage.13 The CataractNational Dataset electronicmulticenter audit of 55 567 cataract surgeries found anassociation between continuous anticoagulant treat-ment and significant increases in minor complicationsof sharp-needle and sub-Tenon cannula local anesthe-sia.12 However, the treatments were not associatedwithpotentially sight-threateningcomplications relatedto local anesthesia or intraoperative hemorrhages.13

Barequet et al.11 report that patients having phacoemul-sification with continuous anticoagulant treatment hadno significant intraoperative bleeding. However, 6% of

- VOL 36, JULY 2010

1118 ANTIPLATELET AND ANTICOAGULANT MEDICATIONS BEFORE CATARACT SURGERY

patients had minor postoperative ocular bleeding thatin all cases resolved within 1 week without affectingvisual acuity. Gainey et al.7 evaluated hemorrhagiccomplications in patients taking warfarin and foundno significant differences between thosewho continuedthe therapy and those who did not.

Ina studybyKumar et al.,13patients onantiplatelet oranticoagulant therapy had significantly more frequentsubconjunctival hemorrhages than those not followinga therapeutic regimen. In addition, Gauba et al.14 foundthat controlled localized bipolar conjunctival cauterybefore sub-Tenon anesthesia injections can significantlyreduce the frequency of subconjunctival hemorrhage,especially in patients taking warfarin or aspirin.

Patients who have atrial fibrillation with coexistingcerebrovascular disease and those with nontissueprosthetic heart valves have a much better prognosiswhen they are on anticoagulant therapy. However,when warfarin therapy is stopped and then restarted,it reverses the anticoagulation effect for unpredictableperiods of time and can expose the patients to a tran-sient, yet dangerous, hypercoagulable state.1–4 Asystematic review by Dunn and Turpie2 found nosignificant difference in the thromboembolic eventrates between different perioperative managementstrategies. Thromboembolic event rates were 0.4%for continuation of anticoagulant treatment, 0.6% fordiscontinuation of treatment without administrationof intravenous heparin, 0.0% for discontinuation oftreatment with administration of intravenous heparin,and 0.6% for discontinuation of treatment with admin-istration of low-molecular-weight heparin. Whenanticoagulant therapy was discontinued, there wasno significant increase in the incidence of thrombo-emobolitic events. However, when an event did occur,it was much more serious and led to a significantincrease in morbidity.

Fewer than half the hospitals in Japan reportcontinuing antiplatelet or anticoagulant therapy untilthe time of surgery.15 However, most members ofCanadian Society of Cataract and Refractive Surgerystated that they did not stopwarfarin or aspirin duringthe perioperative period for cataract surgery.16 Unfor-tunately, the current information cannot be used toconclusively define a best strategy to use in allpatients, therefore limiting its potential use. Morerigorous studies are needed to allow surgeons tomake better decisions for patients who are followingan antiplatelet or anticoagulant regimen.

The current study has several limitations. First, thedesign was nonrandomized and because the samplewas relatively small, the analytical powerwasnot greatenough to detect small differences that might havebeen present. The small sample also precluded the abil-ity to assess safety. A large-scale randomized study

J CATARACT REFRACT SURG

would be required to determine the safety of continu-ous antiplatelet and anticoagulant treatment associ-ated with phacoemulsification and IOL implantation.

Although the sample in each group was small, thisstudy has 2 significant findings. First, the incidenceof subconjunctival hemorrhage at the time of phaco-emulsification was significantly higher in patientswho continued warfarin and/or aspirin therapy thanin those who did not. Second, there were no significantdifferences between the 2 groups in the incidence ofintraoperative or postoperative complications or invisual improvement. To confirm these observations,a studywith a larger population is required. However,the findings in the present study may be of clinical usewhen evaluating the long-term administration ofwarfarin or aspirin in cataract patients.

REFERENCES1. Konstantatos A. Anticoagulation and cataract surgery: a review

of the current literature. Anaesth Intensive Care 2001; 29:11–18

2. Dunn AS, Turpie AGG. Perioperative management of patients

receiving oral anticoagulants; a systematic review. Arch Intern

Med 2003; 163:901–908. Available at: http://archinte.ama-

assn.org/cgi/reprint/163/8/901. Accessed March 3, 2010

3. Jafri SM. Periprocedural thromboprophylaxis in patients receiving

chronic anticoagulation therapy. Am Heart J 2004; 147:3–15.

Available at: http://www.pha.nu.ac.th/apirukw/ambu/uploads/

4D746_PeriprocedureAnticoag.pdf. Accessed March 3, 2010

4. Douketis JD, Berger PB, Dunn AS, Jaffer AK, Spyropoulos AC,

Becker RC, Ansell J. The perioperative management of

antithrombotic therapy: American College of Chest Physicians

Evidence-Based Clinical Practice Guidelines (8th Edition).

Chest 2008; 133(6 suppl):299S–339S. Available at http://chest-

journal.chestpubs.org/content/133/6_suppl/299S.full.pdf.

Accessed March 3, 2010

5. Emery JM, Little JH. Patient selection. In: Emery JH, Little JH,

eds, Phacoemulsification and Aspiration of Cataract; Surgical

Techniques, Complications and Results. St Louis, MO, Mosby,

1979; 45–48

6. Teekhasaenee C, Ritch R. The use of PhEA 34c in trabeculec-

tomy. Ophthalmology 1986; 93:487–491

7. Gainey SP, Robertson DM, Fay W, Ilstrup D. Ocular surgery on

patients receiving long-term warfarin therapy. Am J Ophthalmol

1989; 108:142–146

8. Carter K, Miller KM. Phacoemulsification and lens implantation

in patients treated with aspirin or warfarin. J Cataract Refract

Surg 1998; 24:1361–1364

9. Morris A, Elder MJ. Warfarin therapy and cataract surgery. Clin

Exp Ophthalmol 2000; 28:419–422

10. Katz J, Feldman MA, Bass EB, Lubomski LH, Tielsch JM,

Petty BG, Fleisher LA, Schein OD. Risks and benefits of

anticoagulant and antiplatelet medication use before cataract

surgery; the Study of Medical Testing for Cataract Surgery

Team. Ophthalmology 2003; 110:1784–1788

11. Barequet IS, Sacks D, Priel A, Wasserzug Y, Martinowitz U,

Moisseiev J, Salomon O. Phacoemulsification of cataract in

patients receiving Coumadin therapy: ocular and hematologic

risk assessment. Am J Ophthalmol 2007; 144:719–723

12. Benzimra JD, Johnston RL, Jaycock P, Galloway PH,

Lambert G, Chung AKK, Eke T, Sparrow JM. The Cataract

National Dataset electronic multicentre audit of 55 567

- VOL 36, JULY 2010

1119ANTIPLATELET AND ANTICOAGULANT MEDICATIONS BEFORE CATARACT SURGERY

operations: antiplatelet and anticoagulant medications; the EPR

User Group. Eye 2008; 23:10–16

13. Kumar N, Jivan S, Pulpa V, McLure H. Sub-Tenon’s anesthesia

with aspirin, warfarin, and clopidogrel. J Cataract Refract Surg

2006; 32:1022–1025

14. Gauba V, Saleh GM, Watson K, Chung A. Sub-Tenon anaesthe-

sia: reduction in subconjunctival haemorrhage with controlled

bipolar conjunctival cautery. Eye 2007; 21:1387–1390

J CATARACT REFRACT SURG

15. Yasaka M, Okada Y, Inoue T, Yoshikawa H, Saku M. [Question-

naire survey to investigate correspondence of medical doctors

and dentists in Japan for antithrombotic therapy at surgeries or

biopsy]. [Japanese] Brain Nerve 2007; 59:871–876

16. Ong-Tone L, Paluck EC, Hart-Michell RD. Perioperative use of

warfarin and aspirin in cataract surgery by Canadian Society

of Cataract and Refractive Surgery members: survey. J Cataract

Refract Surg 2005; 31:991–996

- VOL 36, JULY 2010