Outpatient Oral Anticoagulant Management — AnAudit of 82 Patients

N Kakkar, R Kaur, Mary John

Original Article

Objective :
Use of oral anticoagulants for thrombotic diseases has been increasing steadily over the years.Management practices however, are far from uniform. We conducted a retrospective audit among outpatientson oral anticoagulant therapy to assess treatment practices and overall control of anticoagulation.

Methods : Case records of 82 patients who were on anticoagulant therapy for a minimum duration of threemonths were reviewed. Information on pre-therapeutic assessment of patients, therapeutic control andcomplications seen during the course of treatment was recorded.

Results : Case notes of 43 males and 39 females with a mean age of 47.5 ± 14.6 years, on oral anticoagulanttreatment were evaluated. Treatment duration ranged from 3 months to 7 years for a total of 258.7 patienttreatment years. Pre-therapeutic assessment of patients was inadequate with only baseline hematologicaland renal parameters available for most patients. Of a total of 1631 prothrombin time ratios and InternationalNormalized Ratios recorded, only 17.8% were in the therapeutic range with 73% being sub-therapeutic.Sixteen (19.5%) patients had treatment related complications. The number of thrombotic and hemorrhagicevents per 100 patient treatment years was 3.4 and 2.7 respectively.

Conclusions : Pre-therapeutic assessment of patients was inadequate. The overall therapeutic control waspoor with patients in a state of underanticoagulation for most period of anticoagulant treatment. Thecomplication rate was also unacceptably high. There is a need to reassess management practices of patientson long term oral anticoagulation with strict adherence to standard accepted guidelines to make this therapymore effective and safer for patients. ©


Oral anticoagulation although of proven benefit in anumber of disorders can be dangerous withoutcareful monitoring.1 Managing long term oralanticoagulant therapy in patients is a complex taskrequiring carefully timed laboratory testing withappropriate dosage adjustment and prompt diagnosisand management of thromboembolic or hemorrhagiccomplications. Since oral anticoagulant drugs possessa narrow therapeutic index, therapeutic control isdifficult. Alterations in diet, concomitant drug or alcoholintake, intercurrent illness and hereditary or acquiredresistance to anticoagulants can further complicate themanagement of such patients.2 With nearly six millionpatients all over the globe on long term oralanticoagulants, this aspect of medical management hasassumed great importance. Consistent regulation of anticoagulant treatment has become safer since theintroduction of the international normalized ratio(INR).3,4 Quality of management of patients and the co-ordination of anticoagulation services may howeverneed review.

We carried out an audit on oral anticoagulant controlin the out-patient setting of our hospital.

Reader; Lecturer; Professor, Departments of Pathologyand Medicine, Christian Medical College and Hospital,Ludhiana, Punjab.

Received : 6.4.2005; Revised : 14.6.2005; Accepted : 16.8.2005


The study was carried out in the Departments ofPathology and Medicine, Christian Medical College andHospital, Ludhiana. Case records of 82 patients whowere on outpatient oral anticoagulant therapy for aminimum duration of three months were reviewed.Information retrieved pertained to pre-treatmentassessment of patients: documentation of dietary anddrug history, baseline complete blood count, renal, liverfunction tests and the coagulation profile. Therapeuticpractices and control of anticoagulation was evaluatedby the indication of starting therapy, duration of therapy,loading dose given and prothrombin times (PT)/International Normalized Ratios (INRs) done till the last follow up. Also noted was the maintenance of a separatePT/INR record, physician response to INR values abovetherapeutic range, the interval of repeat PT/INR testingand complications encountered during the course oftreatment. In some patients on long term follow up,prothrombin time ratio (PTR) instead of INR was usedto record adequacy of anticoagulant control as the INRmode of reporting in the laboratory was introduced onlyin 1998. For patients with valve replacement andrecurrent deep vein thrombosis, an INR range of 3.5 wasconsidered as the target value, while a range of 2.0-3.0was considered for other indications requiring oralanticoagulants.5


Patient profile : This study included 82 patients (43males and 39 females) on oral anticoagulant therapy fora duration ranging from 3 months to 17 years for a totalof 258.7 patient treatment years. The indications forwhich these patients were started on anticoagulanttherapy are listed in Table 1. The ages of the patientsranged from 14 years to 72 years with a mean age of 47.5± 14.6 years. Over half (45/82) of the patients were aged50 years or above.

Pre-therapeutic assessment : Documentation of dietand drug history along with maintenance of a separatePT/INR record was poor. In all patients, baselinehematological data was available and majority (79/82)had evaluation of renal parameters prior to initiation ofanticoagulant therapy. Coagulation profile and liverfunction tests were infrequently performed. The detailed data on pre-therapeutic assessment of patients is shownin Fig. 1. None of the case records had documentation ofthe desired target INR at initiation of treatment.

Therapeutic control : All but one patient (in whomwarfarin was used) were prescribed acenocoumerol. Inmost (50) patients, a 2 mg loading dose of acenocoumerolwas used. The loading dose in eighteen patients was 4mg, 1 mg in nine patients and 3 mg in four patients. Inthe lone case in which warfarin was used, a loadingdose of 10 mg was given. Sixty three (76.8%) patientswere on alternate strength dosage at some time duringthe course of therapy while the rest were on a stabledosage regimen. A total of 1631 prothrombin time ratio(PTR)/International normalized ratio (INR) estimationswere carried out in patients of the study group in 3105months of oral anticoagulant therapy for an average of0.5 PT/INRs per month.

Majority of the PT ratios and INR values in thepatients were sub-therapeutic with only 17.8% in thetherapeutic range (Fig. 2). The percentage of PTR/INRsin the therapeutic range for each patient throughout thecourse of therapy was poor (Fig. 3). Physicians stoppedthe anticoagulant in response to an INR above thetherapeutic range in 51 patients while the drug dosagewas reduced in 19 patients. In twelve patients, the INRdid not overshoot the therapeutic range.

The frequency of repeat PT/INR requests after a highvalue was highly variable (Fig. 4) with majority of thembeing made on the third day after the high INR valuewas recorded.

Complications : Sixteen (19.5%) patients had treatmentrelated complications of which nine were thromboticwhile seven were hemorrhagic; the overall complicationrate being 6.2/100 patient treatment years. Of the ninepatients with thrombotic events, two had recurrence ofthe initial disease (left ventricular clot and myocardialinfarction), one of whom expired; while the remainingseven developed thrombosis at new sites-cerebrovascular accident (4), peripheral arterialembolism (2) and deep vein thrombosis (1). Prothrombintime was done in six of the nine patients at admission.The prothrombin time ratio was sub-therapeutic in allsix patients (mean=1.3). Among the seven patients withhemorrhagic complications, two patients each hadhemoptysis, bleeding per rectum and bleeding from gumswhile one patient had hematemesis. INR was done onlyin one patient in whom it was grossly deranged (6.84).There was however no major bleed requiringhospitalization or blood transfusions in patients withhemorrhagic complications. There was no mortality inthis group.

Fig. 1 : Shows the pre-therapeutic assessment of patients in the studygroup. Note that except for hematological and renal parameters,other evaluations are inadequate.
Fig. 2 : Shows the prothrombin time ratio (PTR) and InternationalNormalized Ratios (INRs) (n=1631) performed in 82 patients of thestudy group. Note that a vast majority of values are below thetherapeutic range with only less than one-fifth in the therapeuticrange.
Fig. 4 : Shows the frequency of repeat testing of PTR/INR after ahigh value in fifty eight patients. In twelve patients repeat testingwas not done and in another twelve, the PTR/INR did not overshootthe therapeutic range at any time during the course of therapy.

None of the patients had both hemorrhagic andthrombotic complications. Three of the 16 patients whodeveloped complications had two thrombotic events while on follow up.

There were 3.4 thrombotic events per 100 treatmentyears while the hemorrhagic events were 2.7 per 100patient treatment years.


This study has shown that outpatient anticoagulantcontrol in patients was generally poor with inadequatepre-therapeutic assessment, an unacceptably highproportion of subtherapeutic PTR/INR values and ahigh complication rate.Guidelines now exist which can help in standardizingthe use of oral anticoagulants in various clinical settings.5(Table 2).More than half of the patients in our study were abovethe age of 50 years. Factors like poor comprehension ofdosage and precautions to be taken, poor health, limitingambulation and difficulty in frequent laboratory testingin the elderly can make therapeutic control difficult6although few studies have also shown adequateanticoagulant control in the elderly population.7Pretherapeutic assessment of patients was found tobe inadequate with no documentation of relevant dietaryhistory at all while drug history was recorded in onlyfour patients. Of these, the drug dosage was adjusted inone patient after he was started on Rifampicin. Althoughbaseline hematological and renal profile was availablefor most patients, liver functions and coagulation profilewere infrequently done. The often ignored dietalterations8 and concomitant drug use9 can seriouslydestabilize anticoagulant control in patients. Tan et al10audited anticoagulant therapy in inpatients. Just over50% patients in their study had baseline coagulationscreens while 70% had liver function evaluation beforestarting anticoagulant therapy. Similar to our study,baseline hematological and renal parameters were available for most patients. Thirty three percent of APTTsand 58% of INRs in their study were subtherapeutic. Inour study, which pertained only to outpatients, we alsofound an alarming proportion (73%) of PTR/INRs to besub-therapeutic. Azar et al11 looked at the quality controlof oral anticoagulant therapy in 1700 patients followingmyocardial infarction. They reported that therapeuticINRs were difficult to achieve in the first 6 months oftherapy with <60% of INRs within the therapeutic range.After treatment stabilized, the figures increased to 70%in patients on acenocoumerol and 80% forphencoumoron. In contrast, in our study, even patientswith long term follow up (up to 17 years) continued tohave a large proportion of INRs in the sub-therapeuticrange. Studies have shown that only about 50% of INRsin a patient population taking warfarin are within thetherapeutic range.12,13 In the present study, repeat testingafter a high INR value was done in some patients within24-48 hours, an interval too short for clinical assessment.

Thrombotic complications in patients with pooranticoagulant control are expectedly higher. de Vincenteet al14 reported a 3.31 times higher risk of majorthromboembolic complications in patients who had amean INR below the therapeutic range. Alonso et al15assessed therapeutic control and frequency ofcomplications in 136 outpatients on oral anticoagulants.In their study, nearly 70% of all INR determinations werewithin satisfactory range. Frequency of minorhemorrhage was 13.36/100 patient years, majorhemorrhage- 1.08 and thromboembolism was 2.53/100patient years. In contrast, our study showed gross under-anticoagulation status in patients and not surprisinglynearly equal proportions of thrombotic as well as hemorrhagic complications. Also, the fact that none ofthe patients in our study had a major bleed, suggeststhat hemorrhagic complications did not pose asignificant clinical problem.

Although percentage time spent by patients in thetherapeutic range is a better method of assessing controlof anticoagulant therapy, due to lack of computerizedcoagulation records and inconsistent documentation ofthe date of PT/INR testing, we chose to report thepercentage of INRs in the therapeutic range as an indexof therapeutic control in our study.

Oral anticoagulation therapy requires regular testingto assess the degree of anticoagulation achieved. Addedto this, the inherent problems with drugs used and theirinteractions with diet and drugs make it difficult toachieve optimal control. However, newer modalities forachieving anticoagulation, now frequently used in theWest, like point of care self-managed testing systems16and telephonic consultations17 have been veryencouraging and may be the tools of tomorrow to dealwith this vexed issue.


Outpatient anticoagulant control in patients wasinadequate with most patients in a state of underanticoagulation. Adherence to standard guidelines ofmanagement was also not optimal.

Careful pre-therapeutic assessment of patients beforeinitiating anticoagulant therapy, establishment of acentralized anticoagulant clinic, strict adherence tostandard management guidelines with the assistance ofcomputer-aided decision support systems and betterpatient education can help greatly in optimizingtreatment in patients on long term anticoagulants.Routine audit of clinical management should be anintegral part of any anticoagulation service. It canidentify areas of deficit towards which appropriatemeasures can then be taken.

We recommend that a protocol sheet (Appendix I) beused for every patient who is started on anticoagulanttherapy to standardize management of patients.


We thank Mr. Rai Singh and other staff of the MedicalRecords section for assistance in retrieval of the caserecords of the patients included in the study.


1.Shetty HG. Routledge PA. Optimising the dose of oralanticoagulants. Ann Acad Med Singapore 1991;20:161-4.

2.Berrettini M. Anticoagulation clinics: the Italian experience.Haematologica 1997;82:713-7.

Drugs that potentiate the action of oral anticoagulant
Drugs that antagonize the effect of the of oral anticoagulants
Chloral hydrate

Frequency of INR estimation-

a) If patient is given heparin, oral anticoagulant drug is started on the same day. Heparin is continued till INR is in the therapeuticrange for two consecutive days.

b) For rapid oral anticoagulation, give the drug daily for 4 days, then monitor weekly to achieve the therapeutic range and stablecontrol. Once stabilized, the interval of testing can be increased up to 12 weeks.Information booklet given to patient- Yes/No

Verbal information given to patient- Yes/No

PT booklet given to patient- Yes/No

Special precautions-

a) In patients with antiphospholipid syndrome, the baseline coagulation parameters are already deranged. Hence a target INR of 3.5for such patients is recommended.

b) In patients with associated liver disease, lower dosage regimens and more frequent monitoring is required.

3.Hirch J, Poller L. The International Normalized Ratio. ArchIntern Med 1994;154:288-92.

4.Nichols WL, Bowie EJ. Standardization of the prothrombintime for monitoring orally administered anticoagulanttherapy with use of the international normalized ratio system.Mayo Clin Proc 1993;68:897-8.

5.Haemostasis and Thrombosis Task Force for the BritishCommittee for Standards in Haematology. Guidelines onoral anticoagulation: Third edition. Br J Haematol1998;101:374-87.

6.Froom P, Miron E, Barak M. Oral anticoagulants in the elderly.Br J Haematol 2003;120:526-8.

7.Kagansky N, Knobler H, Rimon E, Ozer Z, Levy S. Safety ofanticoagulation therapy in well-informed older patients. AnnIntern Med 2004;164:2044-50.

8.Chow WH, Chow TC, Tse TM, Tai YT, Lee WT.Anticoagulation instability with life-threatening complicationafter dietary modification. Postgrad Med J 1990;66:855-7.

9.Wittlowsky Ak, Boccuzzi SJ, Wogen J, Wygant G, Patel P,Haunch O. Frequency of concurrent use of warfarin withpotentially interacting drugs. Pharmacotherapy2004;24:1668-74.

10.Tan G, Cohen H, Taylor F, Gabbay J. Audit of start ofanticoagulation treatment in inpatients. J Clin Pathol1993;46:67-71.

11.Azar AJ, Deckers JW, Rosendaal FR, van Bergen PF, van derMeer FJ, Jonker JJ, et al. Assessment of therapeutic qualitycontrol in a long-term anticoagulant trial in post-myocardialinfarction patients. Thromb Haemost 1994 ;72:347-51.

12.Rose P. Audit of anticoagulant therapy. J Clin Pathol1996;49:5-9.

13.McCormick D, Gurwitz JH, Goldberg RJ, Becker R, Tate JP,Elwell A, et al. Prevalence and quality of warfarin use forpatients with atrial fibrillation in the long-term care setting.Arch Intern Med 2001;161:2458-63.

14.de Vicente Camara MP, Lucia Cuesta JF, Aguilar Franco C,Solano Bernad V, Serrano Gonzalez, Garcia-Erce JA. Sangre1999;44:469-72.

15.Alonso Roca R, Puche Lopez N, de la Fuente Arriaran MD,Serrano Santos P, Garcia Monterrubio L. Therapeutic qualitycontrol of follow-up of oral anticoagulation in primary care:4-year experience. Aten Primaria 1995;15:555-60.

16.Christensen TD. Self-management of oral anticoagulanttherapy: a review. J Thromb Thrombolysis 2004;18:127-43.

17.Goldberg Y, Meytes D, Shabtai E, Shinron O, Shainberg B,Seligsohn U, Berliner S. Monitoring oral anticoagulant therapyby telephone communication. Blood Coagul Fibrinolysis2005;16:227-30.

18.Bharat V, Das NK, Mohanty B, Jha AC, Chawla SC, Dash Bet al. Reduction of mechanical heart valve thrombosis througha clinical audit. J Heart Valve Dis 2003;12:362-9.


HIV Congress 2006

Jointly organized by AIDS Society of India, BHIVA, EACS, Association of Physicians of India - MumbaiBranch, Indian Academy of HIV and Liver Diseses, Mumbai Districts AIDS Control Society andRajasthan State AIDS Control Society. 10-12th March 2006. Venue : Taj Lands Ends Hotel, Mumbai.

For further details contact : Dr. JK Maniar, Organising Chairperson, 69/2, Walkeshwar Road,Mumbai 400 006.

Tel: 91-22-23680288; Fax: 91-22-22083184

Email: jkmaniar@vsnl.com

Web: hivcongress2006.com

Organising Secretary : Milind Nadkar, E-mail : milindnadkar@yahoo.com



80th National Conference of Indian Medical Association. 27th to 29th December, 2005 at Chennai TradeCentre, Mount Poonamallee Road, Nandambakkam, Chennai - 600 089.

Organizing Secretary : Dr. T Sadagopan, Usha Nursing Home, 23/2, Filter Bed Road, Vellore - 632 001,Tamil Nadu. Tel: 0416-2224878, 2223222; Mob : 9843034878, 9443374878

Scientific Committee Convenor : Dr. S Chandrasekharan, No.4, Jambulingam Street, Nungambakkam,Chennai - 600 034.

Tel: 044-28275832, 28312946; Mob : 9840015023;

Email: schan2004@rediffmail.com