Abbreviated Hospitalization for Deep Venous Thrombosis with the Use of Ardeparin

ORIGINAL INVESTIGATION Abbreviated Hospitalization for Deep Venous Thrombosis With the Use of Ardeparin

Samuel Z. Goldhaber, MD; Ruth B. Morrison, RN, BSN, CVN; Linda L. Diran, BS; Mark A. Creager, MD; Thomas H. Lee, Jr, MD

Background: Ardeparin sodium has recently re- ultrasonography. Evaluation of baseline vs 6-week ceived approval by the Food and Drug Administration venous scans demonstrated that, overall, 31 of the for prophylaxis against venous thromboembolism in 39 ardeparin-treated patients improved, compared patients undergoing elective total knee replacement. with 21 of the 36 patients assigned to receive unfrac- However, this low-molecular-weight heparin has not tionated heparin (P = .05). The 95% confidence in- been previously evaluated in a randomized controlled terval for the difference in improvement was 0.6% trial for treatment of established acute deep venous to 42% in favor of ardeparin. Median charges for arde- thrombosis. parin and unfractionated heparin were $2815 and $6500, respectively (PϽ.001). There were no differ- Methods: The study included patients with ultrasound- ences in bleeding or patient satisfaction between the 2 documented acute symptomatic deep venous thrombo- groups. sis of the legs. They had to be deemed appropriate for discharge home to receive subcutaneous low-molecular- Conclusions: The results of this small preliminary weight heparin. Patients were randomized to receive ar- trial suggest that ardeparin can be administered effec- deparin with a 2-day hospitalization or unfractionated tively and safely to selected patients with acute deep heparin sodium with a 5-day hospitalization. Both groups venous thrombosis and that, with proper nursing and received warfarin sodium. Follow-up ultrasound exami- home services, it can help decrease the duration of nations were undertaken at 6 weeks. hospitalization.

Results: Of the 80 patients enrolled, 75 had follow-up Arch Intern Med. 1998;158:2325-2328

RDEPARIN SODIUM has re- sessed efficacy, safety, hospital charges, and cently received approval patient satisfaction in the 2 treatment by the Food and Drug groups. Administration for prophylaxis against venous RESULTS thromboembolism in patients undergo- A 1-3 ing elective total knee replacement. How- Overall, 80 patients were randomized, 41 ever, this low-molecular-weight heparin to receive ardeparin and 39 to receive has not been previously evaluated in a ran- unfractionated heparin. Baseline charac- domized controlled trial for treatment of teristics were similar in both groups established acute deep venous thrombo- (Table 1). Of the 80 patients, there were sis (DVT). Therefore, in a single-center, no deaths or recurrent DVTs. There was investigator-initiated trial, we tested the 1 symptomatic, hemodynamically stable feasibility of using a shortened hospital stay pulmonary embolism diagnosed by mod- with ardeparin to treat symptomatic pa- erately high-probability lung scan in a patients with acute DVT proved on ultra- tient assigned to receive unfractionated sound. They were randomized to receive heparin. Adverse clinical events are shown either twice-daily subcutaneously in- in Table 2. One patient in each group suf- jected ardeparin with a 2-day hospital- fered a major bleeding complication. The From the Cardiovascular ization or conventional continuous-in- patient assigned to the unfractionated hep- Division, Department of fusion unfractionated heparin sodium arin group had an intraocular hemor- Medicine, Brigham and with a 5-day hospitalization as a bridge to rhage, and the patient who received arde- Women’s Hospital, Harvard oral anticoagulation with warfarin so- parin had undergone thoracic surgery and Medical School, Boston, Mass. dium. In this feasibility study, we as- developed mediastinal bleeding.

ARCH INTERN MED/ VOL 158, NOV 23, 1998 2325

Downloaded From: https://jamanetwork.com/ on 09/28/2021 PATIENTS AND METHODS Patients were given a satisfaction questionnaire to return at the time of their 2-week visit. At the 6-week office visit, they underwent venous ultrasonography. Dur- PATIENTS ing the time from hospital discharge to 6 weeks, international normalized ratios were checked at least once The study included patients 18 years of age or older who weekly. At the 6-week visit, the trial ended and manage- had acute (within 14 days) symptomatic DVT of the legs. ment of warfarin dosing was transferred to the primary They had to be deemed appropriate for discharge home to care physician. receive subcutaneous low-molecular-weight heparin, and A sonographic unit (Acuson 128; Acuson, Mountain DVT had to be documented by ultrasound. Our Human View, Calif) was used to perform venous ultrasound ex- Research Committee approved this trial, and each patient aminations of the lower extremities, with the use of 5- and provided written informed consent. 7-MHz transducers (duplex and color pulsed Doppler ca- Principal exclusion criteria were high-risk DVT in- pacity). Studies were obtained at baseline (before treat- volving 3 proximal veins (eg, common femoral, superfi- ment) and were repeated at 6 weeks. cial femoral, and popliteal veins); pelvic vein thrombosis; Examinations were performed by compressing current symptomatic pulmonary embolism; expected the deep veins of the thigh and the calf in a sequential prolonged hospitalization for other reasons; hemoglobin manner in 1- to 2-cm increments along the common fem- level less than 85 g/L or platelet count less than 100 ϫ 109/L; oral, superficial femoral, popliteal, posterior tibial, pero- intracranial or intraocular surgery, stroke, or internal bleed- neal, gastrocnemius, and soleus veins. Lack of venous ing within 3 months; or weight of 120 kg or more. compressibility with the ultrasound transducer held Randomization was accomplished by calling a central transverse to the artery and vein was interpreted as an computerized service based at Wyeth-Ayerst Research, Phila- abnormal study result and was confirmed with color-flow delphia, Pa. The drug assignment was not blinded. and pulsed-wave Doppler analysis.5 Absent or diminished Doppler flow, lack of respiratory variation, and failure to METHODS augment flow with maneuvers (calf compression) were used to confirm the diagnosis of DVT. One of us (M.A.C.) Patients assigned to the ardeparin sodium group received was responsible for adjudicating the comparison of base- 130 anti-Xa U/kg subcutaneously twice daily and were hos- line and 6-week follow-up ultrasound scans. He had no pitalized for 36 to 48 hours. Ardeparin was administered knowledge of the randomization assignment. Improve- (usually by the patient) for 5 to 15 days, while warfarin so- ment, no change, and worsening of DVT were defined by dium, 7.5 to 10 mg, was initiated and titrated to a thera- regression, no change, or extension, respectively, of visu- peutic and stable international normalized ratio of 2.0 to alized thrombus. 3.0. During ardeparin treatment, daily international nor- We previously found that duplex ultrasonography, malized ratios were obtained, and, after discharge, a study when compared with venography, is a reliable technique nurse (R.B.M.) telephoned the patient daily to discuss war- for the detection of suspected infrapopliteal DVT.6 In a se- farin dosing as well as overall progress. Patients assigned ries of 30 symptomatic patients undergoing both contrast to the unfractionated heparin sodium group received a 5000- venography of the calf veins and ultrasonography, 7 had to 7500-unit bolus followed by a continuous intravenous venographically documented isolated calf DVT, and all 7 infusion administered for 5 days or more to achieve a tar- cases were detected by ultrasonography. get partial thromboplastin time of 1.5 to 2.5 times the up- per limit of the control. Titration of unfractionated hepa- STATISTICAL ANALYSIS rin was guided by the Cruickshank et al nomogram4 and was discontinued after 5 days as long as the target inter- Data were entered into TRUE EPISTAT files and analyzed national normalized ratio had been achieved. by means of TRUE EPISTAT software (Epistat Services, Ri- Hospital charges were tracked and computed by the chardson, Tex). Student t test was used to examine differ- Computer Assisted Hospitalization Analysis for the Study ences among continuous variables with normal distribu- of Efficacy at Brigham and Women’s Hospital, Boston, Mass. tions. The Wilcoxon rank sum test was used when These consisted of charges for room, board, laboratory tests, continuous variables failed tests for normality. Differ- medications, and imaging tests. ences among discontinuous variables were analyzed with All patients were discharged home with vascular com- Epi Info software (version 5.01b; Centers for Disease Con- pression stockings and were scheduled to return for 2- trol and Prevention, Epidemiology Program Office, At- week and 6-week office visits. We instructed all patients lanta, Ga) by means of ␹2 with the Yates correction, ex- about signs and symptoms of recurrent venous thrombo- cept with expected cell values less than 5, in which case embolism. We provided them with 24-hour emergency tele- Fisher exact 2-tailed test was used. Confidence intervals phone numbers. We also suggested to patients in the ar- were calculated with the use of confidence interval analy- deparin group that they not return to their usual activity sis software (Confidence Inverview Analysis, version 1.0; level until ardeparin was discontinued. British Medical Journal, London, England).

The principal efficacy end point was the change on proved, compared with 21 (58%) of the 36 patients as- the 6-week ultrasound scan compared with baseline signed to the unfractionated heparin group (P = .05). The (Table 3). Fifteen ardeparin-treated patients com- 95% confidence interval for the difference in improve- pared with 8 patients who received unfractionated hep- ment was 0.6% to 42% in favor of ardeparin. arin had previous venous thromboembolism (PϽ.001). Hospital charges were greater in the unfraction- Overall, 31 (79%) of the 39 ardeparin-treated patients im- ated heparin group, which averaged 5.7 days of hospi-

ARCH INTERN MED/ VOL 158, NOV 23, 1998 2326

Downloaded From: https://jamanetwork.com/ on 09/28/2021 40 000 Table 1. Baseline Characteristics*

35 000 Treatment Group

Ardeparin Sodium Unfractionated Heparin 30 000 Variable (n = 41) Sodium (n = 39) Age, mean ± SD, y 52.7 ± 15.9 55.3 ± 15.3 Sex, No. M/F 20:21 23:16 25 000 Proximal DVT, No. 20 16 Isolated CVT, No. 21 23 20 000 Previous DVT or PE, No. 15 8 Total Charges, $ Total *DVT indicates deep venous thrombosis; CVT, calf vein thrombosis; 15 000 and PE, pulmonary embolus.

10 000

Table 2. Adverse Clinical Events* 5000

Treatment Group 0 Ardeparin Sodium Unfractionated Heparin Ardeparin Unfractionated Heparin (n = 41) Sodium (n = 39) Scatterplot for hospital charges among patients assigned to receive DVT/PE, No. 0:0 0:1 ardeparin sodium and unfractionated heparin sodium. Charges were much Major bleeding lower for patients in the ardeparin group ( PϽ.001). The charges averaged No. of patients 1 1 (±SD) $4101 ± $3616 for ardeparin vs $11 965 ± $23 670 for unfractionated 95% CI, % 0.07-13.00 0.07-14.00 heparin. Median charges were $2815 and $6500, respectively. Minor bleeding No. of patients 7 9 95% CI, % 7.20-32.00 11.00-39.00 Table 5. Patient Satisfaction Regarding Length of Stay (LOS)

*DVT indicates deep venous thrombosis; PE, pulmonary embolus; Treatment Group, No. of Patients and CI, confidence interval. Ardeparin Sodium Unfractionated Heparin Perception of LOS (n = 38) Sodium (n = 36) Table 3. Change in Ultrasound Findings From Baseline A lot shorter than 10 to 6 Weeks* needed A little shorter than 51 needed Treatment Group, No. of Patients About right 30 27 Ardeparin Sodium Unfractionated Heparin A little longer than 26 Findings (n = 39) Sodium (n = 36) needed A lot longer than 02 Improved 31 21 needed Same 8 14 Worse 0 1

*P = .05 for improvement with ardeparin compared with unfractionated heparin. According to a patient satisfaction scoring system in which 1 indicates excellent; 2, very good; 3, good; 4, fair; and 5, poor, both groups reported an average score Table 4. Length of Stay of 1.3 for the care that they received. At 2 weeks, there was no difference between ardeparin and unfraction- Length of Stay, d, by Treatment Group ated heparin treatment with respect to activities of daily living. Regarding length of stay, 5 ardeparin-treated pa- Ardeparin Sodium Unfractionated Heparin tients thought their hospital stay was “a little shorter than (n = 41) Sodium (n = 39) needed,” whereas 6 patients treated with unfractionated Mean 2.2 5.7 heparin thought their hospital stay was “a little longer Median 2.0 5.0 Table 5 Range 1-7 4-9 than needed” ( ). First quartile range 1-2 4-5 Third quartile range 2-2 5-5 COMMENT Fourth quartile range 2-7 5-9 This trial demonstrates the probability that among a rela- tively small group of properly selected patients with DVT, talization, compared with the ardeparin group, which av- ardeparin administration can be used in lieu of unfrac- eraged 2.2 days of hospitalization (Table 4). For example, tionated heparin to shorten the hospitalization period and the median charges were $6500 and $2815, respectively decrease hospital charges. Efficacy, in terms of throm- (Figure). bus resolution, was greater with ardeparin than with un-

ARCH INTERN MED/ VOL 158, NOV 23, 1998 2327

Downloaded From: https://jamanetwork.com/ on 09/28/2021 fractionated heparin, and safety and patient satisfaction services, it can be used to help decrease the duration of were similar with ardeparin and unfractionated hepa- hospitalization. However, a larger and more definitive rin. The principal difference between the 2 manage- study should be undertaken to confirm our findings. ment strategies in these groups was the dramatically de- creased hospital charges among ardeparin-treated patients. Accepted for publication March 5, 1998. It is important to note that the dose of ardeparin so- This study was supported in part by a grant from Wyeth- dium, 130 anti-Xa U/kg, far exceeds the dose of 50 anti-Xa Ayerst Research, Philadelphia, Pa. U/kg recently approved for DVT prophylaxis among pa- Reprints: Samuel Z. Goldhaber, MD, Cardiovascular tients undergoing total knee replacement. Division, Brigham and Women’s Hospital, 75 Francis St, Bos- Other low-molecular-weight heparins have been ton, MA 02115 (e-mail: [email protected]). demonstrated to be effective in the management of DVT. These include reviparin sodium,7 enoxaparin sodium,8 nadroparin calcium,9 tinzaparin sodium,10 and daltepa- REFERENCES rin sodium.11 Enoxaparin and nadroparin were specifi- cally used to test the strategy of an abbreviated hospital- 1. RD Heparin Arthroplasty Group. RD heparin compared with warfarin for preven- ization or, in some instances, completely outpatient DVT tion of venous thromboembolic disease following total hip or knee arthroplasty. management. However, each low-molecular-weight hep- J Bone Joint Surg Am. 1994;76:1174-1185. 12 2. Levine MN, Gent M, Hirsh J, et al. Ardeparin (low-molecular-weight heparin) vs arin has special biochemical characteristics. For ex- graduated compression stockings for the prevention of venous thromboembo- ample, ardeparin is prepared by peroxidative depolymer- lism: a randomized trial in patients undergoing knee surgery. Arch Intern Med. ization; it has an average molecular weight of 6000 daltons 1996;156:851-856. and an anti-Xa to anti-IIa ratio of 2.0. In contrast, enoxa- 3. Heit JA, Berkowitz SD, Bona R, et al. Efficacy and safety of low molecular weight parin is prepared by benzylation and alkaline depoly- heparin (ardeparin sodium) compared to warfarin for the prevention of venous thromboembolism after total knee replacement surgery: a double-blind, dose- merization. It has an average molecular weight of 4200 ranging study. Thromb Haemost. 1997;77:32-38. daltons and an anti-Xa to anti-IIa ratio of 3.8. 4. Cruickshank MK, Levine MN, Hirsh J, Roberts R, Siguenza M. A standard hep- This study is limited by small sample size. To plan arin nomogram for the management of heparin therapy. Arch Intern Med. 1991; this trial with the goal of 95% confidence and 80% power 151:333-337. 5. Polak JF. Peripheral Vascular Sonography: A Practical Guide. Baltimore, Md: to detect a 20% improvement in efficacy among ardeparin- Williams & Wilkins; 1992:162-166. treated patients, based on the observed 58% improve- 6. Simons GR, Skibo LR, Polak JF, Creager MA, Klapec-Fay JM, Goldhaber SZ. Util- ment rate in the unfractionated heparin group, a sample ity of ultrasonography in suspected symptomatic isolated calf deep venous throm- size of about 400 patients would have been required. bosis. Am J Med. 1995;99:43-47. The present study extends the beneficial results of 7. The Columbus Investigators. Low-molecular-weight heparin in the treatment of patients with venous thromboembolism. N Engl J Med. 1997;337:657-662. ardeparin for orthopedic surgical prophylaxis to the treat- 8. Levine M, Gent M, Hirsh J, et al. A comparison of low-molecular-weight heparin ment of established acute DVT. In a dose more than 2.5 administered primarily at home with unfractionated heparin administered in the times greater than that used for prophylaxis, ardeparin hospital for proximal deep-vein thrombosis. N Engl J Med. 1996;334:677-681. was demonstrated to be more effective and as safe as un- 9. Koopman MMW, Prandoni P, Piovella F, et al. Treatment of venous thrombosis with intravenous unfractionated heparin administered in the hospital as com- fractionated heparin. The use of ardeparin permitted pared with subcutaneous low-molecular-weight heparin administered at home. implementation of an early discharge strategy, which con- N Engl J Med. 1996;334:682-687. served the resources associated with 3 incremental days 10. Hull RD, Raskob GE, Pineo GF, et al. Subcutaneous low-molecular-weight hep- of hospitalization per patient. Importantly, this strategy arin compared with continuous intravenous heparin in the treatment of proximal- was instituted without a decrement in overall patient sat- vein thrombosis. N Engl J Med. 1992;326:975-982. 11. Flessinger JN, Lopez-Fernandez M, Gatterer E, et al. Once-daily subcutaneous isfaction. In summary, this trial suggests that ardeparin dalteparin, a low molecular weight heparin, for the initial treatment of acute deep can be administered effectively and safely to selected pa- vein thrombosis. Thromb Haemost. 1996;76:195-199. tients with acute DVT. With proper nursing and home 12. Weitz JI. Low-molecular-weight heparins. N Engl J Med. 1997;337:688-698.

ARCH INTERN MED/ VOL 158, NOV 23, 1998 2328

Downloaded From: https://jamanetwork.com/ on 09/28/2021