Cost–Effectiveness of Rivaroxaban Versus Heparins for Prevention of Venous Thromboembolism After Total Hip or Knee Surgery in Sweden

Lars Ryttberg; Alex Diamantopoulos; Fiona Forster; Michael Lees; Anina Fraschke; Ingela Björholt

Expert Rev Pharmacoeconomics Outcomes Res. 2011;11(5):601-615.

Five-year View

Rivaroxaban, the direct factor Xa inhibitor apixaban and the direct thrombin inhibitor dabigatran are the most advanced new oral anticoagulants in development. They all have characteristics that simplify their use and overcome the drawbacks associated with the established anticoagulants. These include an oral route of administration, a rapid onset and offset of action, a wide therapeutic window, no requirement for routine coagulation monitoring, and minimal food or drug interactions. The number of hip replacement procedures conducted in Sweden has risen from only six operations in 1976 to 14,105 in 2008, with an increase of almost 4000 operations per year in Sweden between 1998 and 2008.^[104] A similar trend has occurred in knee replacement surgery in Sweden. In the year 2000, it was estimated that, due to an aging population, the number of knee replacement operations would increase by 36% to 7580 operations per year by 2030.^[105] However, this level was reached by 2002 and has since continued to rise. These increases will impose a burden on the Swedish healthcare system. The adoption of rivaroxaban instead of LMWH for thromboprophylaxis after THR and TKR should substantially reduce costs due to events avoided, with the additional result of a substantial improvement in the QoL of these patient groups.

1 2 3 4 5 6 7 8 9 10

Next Section

Acknowledgements
The authors would like to acknowledge Chris Thomas, who provided editorial support.

Financial & competing interests disclosure
Lars Ryttberg has received honoraria from Bayer Pharma AG. Alex Diamantopoulos provided consulting services and is an employee of a private consultancy that provides services to the manufacturer of rivaroxaban. Fiona Forster provided consulting services to Bayer Pharma AG. Michael Lees was an employee of Bayer Plc. Anina Fraschke is an employee of Bayer AB. Ingela Björholt provided consulting services to Bayer Pharma AG. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
Editorial support was provided by Chris Thomas. Funding for this was provided by Bayer HealthCare and Johnson & Johnson Pharmaceutical Research & Development, L.L.C.

Cite this: Cost–Effectiveness of Rivaroxaban Versus Heparins for Prevention of Venous Thromboembolism After Total Hip or Knee Surgery in Sweden - Medscape - Oct 01, 2011.

Table 1. Prophylaxis module: event probabilities.
Event	Rivaroxaban	LMWH
Kakkar et al. (2008) for THR [3]
VTE	0.0197	0.0937
Symptomatic VTE	0.0025	0.0124
Nonfatal PE	0.0012	0.0012^†
Fatal PE	0	0^†
Major bleeding	0.0008	0.0008^†
Lassen et al. (2008) for TKR[4]
VTE	0.0959	0.1880
Symptomatic VTE	0.0067	0.0196
Nonfatal PE	0	0^†
Fatal PE	0	0^†
Major bleeding	0.0057	0.0057^†

^†If the results of the clinical trial do not show any statistically significant difference between the two arms, the base case analysis assumes parity between the two comparators. This assumption is tested in the sensitivity analysis.
LMWH: Low-molecular-weight heparin; PE: Pulmonary embolism; THR: Total hip replacement; TKR: Total knee replacement; VTE: Venous thromboembolism.

Table 2. Utility values for modeled events.
Health state	THR	TKR	Ref.
Prophylaxis-related bleeding	0.531	0.532	[20,32]
No venous thromboembolic event	0.805	0.807	[20]
Asymptomatic VTE (assumption [same as no VTE])	0.805	0.807
Symptomatic DVT	0.676	0.678	[20,28]
PE	0.612	0.613	[20,28]
PTS	0.754	0.739	[20,32]
CTPH	0.581	0.569	[20,33]
Recurrent DVT	0.721	0.706	[20,28]
Long-term utility (no VTE)	0.858	0.841	[20]
Death	0	0

CTPH: Chronic thromboembolic pulmonary hypertension; DVT: Deep vein thrombosis; PE: Pulmonary embolism; PTS: Post-thrombotic syndrome; THR: Total hip replacement; TKR: Total knee replacement; VTE: Venous thromboembolism.

Table 3. Unit cost items, treatment costs and healthcare resource utilization.
Treatment, event or item	Resource use	Cost (SEK)	Assumption	Ref.
Prophylaxis drug cost
Rivaroxaban 10 mg per day	In-hospital	49.76 per day		[Bayer AB, Internal Price List, Data on file]
	Post-discharge (THR)	50.85 per day	Assumed purchase of 30-pack size	[106]
	Post-discharge (TKR)	54.00 per day	Assumed purchase of 10-pack size	[106]
Enoxaparin 40 mg per day	In-hospital	31.43 per day	Assumed biggest and cheapest pack size of 50 units	[102]
	Post-discharge (THR)	34.64 per day	Average of 50- and 10-pack size
	Post-discharge (TKR)	37.85 per day	Assumed purchase of 10-pack size
Dalteparin 5000 units per day	In-hospital	27.81 per day	Assumed biggest and cheapest pack size of 100 units	[102]
	Post-discharge (THR)	35.53 per day	Average of 100- and 10-pack size
	Post-discharge (TKR)	43.25 per day	Assumed purchase of 10-pack size
Prophylaxis-related administration & monitoring
Rivaroxaban	No administration or monitoring required	0.00
Enoxaparin/dalteparin	10% require nurse assistance for outpatient administration in patients' homes for 3 days. Cost per visit: SEK515	51.50		[7,21,22]
Enoxaparin/dalteparin	24% of patients require one additional hospital day at SEK1279 per day	307.07		[Ambring A, Björholt I, Data on file]
Prophylaxis-related bleeding
Bleeding		5545.48^†		[6]
VTE diagnosis
PE diagnosis		4192.11^†		[6]
DVT diagnosis		3626.73^†		[6]
DVT treatment
Inpatient treatment (THR)		20,444.42^†		[6]
Inpatient treatment (TKR)		23,490.99^†		[6]
Outpatient treatment (THR)		12,123.24^†		[6]
Outpatient treatment (TKR)		11,452.46^†		[6]
PE treatment
Inpatient treatment (THR)		24,397.17^†		[6]
Inpatient treatment (TKR)		25,545.67^†		[6]
Outpatient treatment (THR)		29,865.83^†		[6]
Outpatient treatment (TKR)		30,976.90^†		[6]
Long-term complications
Recurrent VTE (THR)Recurrent VTE (TKR)		12,123.2411,452.46	Assumption (same cost as treating DVT post-discharge)	[6]
PTS diagnosis		8682.67^†		[6]
PTS treatment (annual)		6512.74^†		[6]
CTPH diagnosis		8470.32^†		[23]
CTPH treatment		42,236.13^†		[24]

^†Converted to SEK and inflated to 2008 prices.
1 EUR = 9.56084 SEK; 1 SEK = 0.104593 EUR.
CTPH: Chronic thromboembolic pulmonary hypertension; DVT: Deep vein thrombosis; EUR: Euros; PE: Pulmonary embolism; PTS: Post-thrombotic syndrome; SEK: Swedish crowns; THR: Total hip replacement; TKR: Total knee replacement; VTE: Venous thromboembolism.

Table 4. Probabilistic sensitivity analysis.
Variable	Distribution	Notes
Costs
Cost for one additional preoperative hospital night (enoxaparin and tinzaparin)	Gamma (μ = 307.07; 95% CI: 214.95–399.20)	Value used in deterministic model ± 30% used as the limits for the 95% CI
Nurse cost for injection administration	Gamma (μ = 51.50; 95% CI: 36.05–66.96)	Value used in deterministic model ± 30% used as the limits for the 95% CI
Major bleeding event (nonfatal)	Gamma (μ = 5545.48; 95% CI: 3881.84–7209.13)	Value used in deterministic model ± 30% used as the limits for the 95% CI. Same value used for fatal major bleeding event
Testing for DVT	Gamma (μ = 3626.73; 95% CI: 2538.71–4714.75)	Value used in deterministic model ± 30% used as the limits for the 95% CI
Testing for PE	Gamma (μ = 4192.11; 95% CI: 2934.48–5449.74)	Value used in deterministic model ± 30% used as the limits for the 95% CI
Treating DVT during hospitalization	Gamma (μ = 20,444.44; 95% CI: 14,311.09–26,577.74)	Value used in deterministic model ± 30% used as the limits for the 95% CI
Treating PE during hospitalization	Gamma (μ = 24,397.17; 95% CI: 17,080.02–31,716.33)	Value used in deterministic model ± 30% used as the limits for the 95% CI
Treating DVT post-discharge	Gamma (μ = 12,123.24; 95% CI: 8486.26–15,760.20)	Value used in deterministic model ± 30% used as the limits for the 95% CI. Treatment costs for recurrent DVT use this value
Treating PE post-discharge	Gamma (μ = 29,865.83; 95% CI: 20,906.08–38,825.58)	Value used in deterministic model ± 30% used as the limits for the 95% CI
Diagnosis of PTS	Gamma (μ = 8682.67; 95% CI: 6077.87–11,287.47)	Value used in deterministic model ± 30% used as the limits for the 95% CI
Treatment of PTS	Gamma (μ = 6512.72; 95% CI: 4558.92–8466.57)	Value used in deterministic model ± 30% used as the limits for the 95% CI
Diagnosis of CTPH	Gamma (μ = 8470.32; 95% CI: 6246.81–10,544.89)	Lower and upper quartile values used as the limits of the 95% CI
Treatment of CTPH	Gamma (μ = 42,236.13; 95% CI: 29,565.29–54,906.97)	Value used in deterministic model ± 30% used as the limits for the 95% CI
Utility values
THR – no venous thromboembolic event (utility after operation)	Beta (α = 1718; β = 416)	Räsänen et al. reported SD [20] All venous thromboembolic events in the THR population linked to this variable. Utility of asymptomatic VTE takes on this value
TKR – no venous thromboembolic event (utility after operation)	Beta (α = 1496; β = 357)	Räsänen et al. reported SD [20] All venous thromboembolic events in the TKR population linked to this variable. Utility of asymptomatic VTE takes on this value
Prophylaxis-related bleeding	Beta (α = 80; β = 41)	Lenert et al. reported 95% CI with scaled-down SE [32] All venous thromboembolic events have equal or higher utility values imposed in relation to this variable. Symptomatic VTE and PE utility linked to this variable
PTS	Beta (α = 25; β = 3)	Lenert et al. reported CI (scaled-down SE) [32]
Recurrent VTE	Beta (α = 51; β = 0.4)	Mean PTS utilities adjusted for the proportion with mild and severe PTS. Sources: Kahn and Ginsberg [16] and Lenert et al. [32]. Lower limit based on the VAS values and upper limit based on SG values reported in Lenert et al. [32]
CTPH	Beta (α = 6; β =3)	Data from Highland et al. [33]. Assume functional class I for higher limit and functional class III for lower limit for 95% CI
Event probabilities: prophylaxis period
Prophylaxis-related major bleeding	Beta (α = 15.32; β = 5079.84)	For the comparator, the CI for the relative risk is translated into a CI for the event probability, which is used to estimate the SE. If the result of the direct or indirect comparison is not statistically significant, the lower limit of 95% CI is 50% of the mean and the upper limit is 150% of the mean
Venous thromboembolic event	Beta (α = 14.80; β = 411.36)	For the comparator, the CI for the relative risk is translated into a CI for the event probability, which is used to estimate the SE. If the result of the direct or indirect comparison is not statistically significant, the lower limit of 95% CI is 50% of the mean and the upper limit is 150% of the mean. Symptomatic VTE and nonfatal PE linked to venous thromboembolic event
Proportion of patients with a false-positive VTE test: DVT	Beta (α = 6.60; β = 57.46)	The parameters for the beta distribution (α, β) were estimated after assuming the following percentiles: 25% 0.075, 50% mean, 75% 0.125
Proportion of patients with a false-positive VTE test: PE	Beta (α = 7.15; β = 338.42)	The parameters for the beta distribution (α, β) were estimated after assuming the following percentiles: 25% 0.015, 50% mean, 75% 0.025
Probability of death following major bleeding	Beta (α = 113.40; β = 14,099.41)	The parameters for the beta distribution (α, β) were estimated after assuming the following percentiles: 25% 0.0075, 50% mean, 75% 0.0085
Event probabilities: post-prophylaxis module
Risk of VTE after RECORD	Beta (α = 95.99; β = 13,001.52)	Assume SE, which varies depending on the treatment population Risk of PE after RECORD linked to this variable
Risk of asymptomatic VTE developing into symptomatic VTE	Beta (α = 5.89; β = 22.54)	Assume SE, which varies depending on the treatment population Risk of PE after RECORD linked to this variable Probability of asymptomatic VTE developing into PE linked to this variable
Event probabilities: long-term complications period
Probability of developing PTS: year 1	Beta (α = 30.29; β = 137.13)	The parameters for the beta distribution (α, β) were estimated after assuming the following percentiles: 25% 0.16, 50% mean, 75% 0.20. All PTS probabilities for following years linked to this variable
Probability of recurrent VTE: year 1	Beta (α = 33.57; β = 337.06)	The parameters for the beta distribution (α, β) were estimated after assuming the following percentiles: 25% 0.01, 50% mean, 75% 0.1. All PTS probabilities for years 2–5 are linked to this variable
Probability of developing CTPH: year 1	Beta (α = 6.18; β = 193.16)	Data from Pengo et al. [17] Probability in year 2 linked to this variable. The probability of developing CTPH at years 3–5 is not sampled; the parameter is assumed to be zero as in the base case analysis

CI: Confidence interval; CTPH: Chronic thromboembolic pulmonary hypertension; DVT: Deep vein thrombosis; PE: Pulmonary embolism; PTS: Post-thrombotic syndrome; RECORD: Regulation of Coagulation in Major Orthopedic Surgery Reducing the Risk of DVT and PE; SD: Standard deviation; SE: Standard error; SG: Standard gamble; THR: Total hip replacement; TKR: Total knee replacement; VAS: Visual analog scale; VTE: Venous thromboembolism.

Table 5. Base case results: rivaroxaban versus enoxaparin and dalteparin (total hip replacement).
Parameter	Rivaroxaban	LMWH	Incremental versus LMWH
Rivaroxaban versus enoxaparin
Cost (SEK)	2432	2313	119
QALY	3.8671	3.8630	0.0040
Symptomatic venous thromboembolic events	0.0079	0.0381	-0.0303
Incremental cost per QALY gained (SEK)			29,378
Incremental cost per symptomatic VTE avoided (SEK)			3929
Rivaroxaban versus dalteparin
Cost (SEK)	2432	2288	143
QALY	3.8671	3.8630	0.0040
Symptomatic venous thromboembolic events	0.0079	0.0381	-0.0303
Incremental cost per QALY gained (SEK)			35,436
Incremental cost per symptomatic VTE avoided (SEK)			4739

1 EUR = 9.56084 SEK; 1 SEK = 0.104593 EUR.
EUR: Euro; LMWH: Low-molecular-weight heparin; QALY: Quality-adjusted life-year; SEK: Swedish crowns; VTE: Venous thromboembolism.

Table 6. Base case results: rivaroxaban versus enoxaparin and dalteparin (total knee replacement).
Parameter	Rivaroxaban	LMWH	Incremental versus LMWH
Rivaroxaban versus enoxaparin
Cost (SEK)	1457	2329	-873
QALY	3.8072	3.8043	0.0029
Symptomatic venous thromboembolic events	0.0143	0.0362	-0.0219
Incremental cost per QALY gained (SEK)			Rivaroxaban dominates
Incremental cost per symptomatic VTE avoided (SEK)			Rivaroxaban dominates
Rivaroxaban versus dalteparin
Cost (SEK)	1457	2336	-880
QALY	3.8072	3.8043	0.0029
Symptomatic venous thromboembolic events	0.0143	0.0362	-0.0219
Incremental cost per QALY gained (SEK)			Rivaroxaban dominates
Incremental cost per symptomatic VTE avoided (SEK)			Rivaroxaban dominates

1 EUR = 9.56084 SEK; 1 SEK = 0.104593 EUR.
EUR: Euro; LMWH: Low-molecular-weight heparin; QALY: Quality-adjusted life-year; SEK: Swedish crowns; VTE: Venous thromboembolism.

Table 7. Sensitivity aanalysis: total cost and incremental cost per quality-adjusted life-year gained (SEK) in total hip replacement.
Scenario	Cost (SEK)			Effect (QALY)		Inc. cost, SEK (enoxaparin)	Inc. cost, SEK (dalteparin)	Inc. effect	ICER (enoxaparin)	ICER (dalteparin)
Scenario	Rivaroxaban	Enoxaparin	Dalteparin	Rivaroxaban	LMWH	Inc. cost, SEK (enoxaparin)	Inc. cost, SEK (dalteparin)	Inc. effect	ICER (enoxaparin)	ICER (dalteparin)
Base case	2432	2313	2288	3.8671	3.8630	119	143	0.0040	29,378	35,436
Remove prophylaxis cost	661	1427	1427	3.8671	3.8630	-766	-766	0.0040	Rivaroxaban dominates	Rivaroxaban dominates
Discount rates (costs: 0%; effects: 0%)	2436	2335	2310	4.0993	4.0951	102	126	0.0042	24,005	29,794
Increase the cost of 1 night in hospital by 30%	2432	2405	2380	3.8671	3.8630	27	51	0.0040	6608	12,667
Decrease the cost of 1 night in hospital by 30%	2432	2221	2196	3.8671	3.8630	211	235	0.0040	52,148	58,207
Increase cost of nurse administration by 30%	2432	2359	2335	3.8671	3.8630	73	97	0.0040	17,921	23,979
Decrease cost of nurse administration by 30%	2432	2266	2242	3.8671	3.8630	165	190	0.0040	40,836	46,894
Increase cost of diagnosis and treatment of DVT and PE by 30%	2610	2650	2626	3.8671	3.8630	-41	-16	0.0040	Rivaroxaban dominates	Rivaroxaban dominates
Decrease cost of diagnosis and treatment of DVT and PE by 30%	2254	1975	1951	3.8671	3.8630	278	303	0.0040	68,815	74,873
Increase cost of long-term complications by 30%	2457	2435	2410	3.8671	3.8630	22	47	0.0040	5514	11,572
Decrease cost of long-term complications by 30%	2406	2191	2166	3.8671	3.8630	215	240	0.0040	53,243	59,301
Increase value of all utilities by 15%	2432	2313	2288	4.4475	4.4441	119	143	0.0034	34,711	41,869
Decrease value of all utilities by 15%	2432	2313	2288	3.2868	3.2824	119	143	0.0043	27,415	33,069

1 EUR = 9.56084 SEK; 1 SEK = 0.104593 EUR.
DVT: Deep vein thrombosis; EUR: Euros; ICER: Incremental cost–effectiveness ratio; Inc.: Incremental; LMWH: Low-molecular-weight heparin; PE: Pulmonary embolism; QALY: Quality-adjusted life-year; SEK: Swedish crowns.

Table 8. Sensitivity analysis: total cost and incremental cost per quality-adjusted life-year gained (SEK) in total knee replacement.
Scenario	Cost (SEK)			Effect (QALY)		Inc. cost, SEK (enoxaparin)	Inc. cost, SEK (dalteparin)	Inc. effect	ICER (enoxaparin)	ICER (dalteparin)
Scenario	Rivaroxaban	Enoxaparin	Dalteparin	Rivaroxaban	LMWH	Inc. cost, SEK (enoxaparin)	Inc. cost, SEK (dalteparin)	Inc. effect	ICER (enoxaparin)	ICER (dalteparin)
Base case	1457	2329	2336	3.8072	3.8043	-873	-880	0.0029	Rivaroxaban dominates
Remove prophylaxis cost	838	1421	1421	3.8072	3.8043	-582	-582	0.0029	Rivaroxaban dominates
Discount rates (costs: 0%; effects: 0%)	1465	2350	2357	4.0348	4.0318	-885	-892	0.0030	Rivaroxaban dominates
Increase the cost of 1 night in hospital by 30%	1457	2421	2429	3.8072	3.8043	-965	-972	0.0029	Rivaroxaban dominates
Decrease the cost of 1 night in hospital by 30%	1457	2237	2244	3.8072	3.8043	-780	-787	0.0029	Rivaroxaban dominates
Increase cost of nurse administration by 30%	1457	2376	2383	3.8072	3.8043	-919	-926	0.0029	Rivaroxaban dominates
Decrease cost of nurse administration by 30%	1457	2283	2290	3.8072	3.8043	-826	-833	0.0029	Rivaroxaban dominates
Increase cost of diagnosis and treatment of DVT and PE by 30%	1666	2662	2669	3.8072	3.8043	-997	-1004	0.0029	Rivaroxaban dominates
Decrease cost of diagnosis and treatment of DVT and PE by 30%	1248	1996	2004	3.8072	3.8043	-748	-755	0.0029	Rivaroxaban dominates
Increase cost of long-term complications by 30%	1502	2443	2450	3.8072	3.8043	-941	-948	0.0029	Rivaroxaban dominates
Decrease cost of long-term complications by 30%	1412	2216	2223	3.8072	3.8043	-804	-811	0.0029	Rivaroxaban dominates
Increase value of all utilities by 15%	1457	2329	2336	4.3789	4.3764	-873	-880	0.0031	Rivaroxaban dominates
Decrease value of all utilities by 15%	1457	2329	2336	3.2357	3.2326	-873	-880	0.0027	Rivaroxaban dominates

Comments

Commenting is limited to medical professionals. To comment please Log-in.

Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.