Short-term Cost Comparison of Systemic Heparin Therapy vs. Catheter Directed Thrombolysis for the Treatment of Massive and Submassive Pulmonary Embolism with Long-Term Chronic Pulmonary Hypertension Cost Model

Abstract

Introduction:

Pulmonary embolism (PE) is a significant disease process that affects an estimated 117 cases per 100,000 person-years. Chronic pulmonary hypertension (CPH) is a long-term complication associated with acute PE which has a significant cost to treat, ranging from $98,000–117,000.

Methods:

A retrospective chart review of 341 patients from January 2011 to November 2018 who presented with massive or submassive PE and were treated with either systemic heparin therapy or catheter directed thrombolysis (CDT). The results of the short-term cost analysis and pulmonary hypertension rates from data collected was then used in a long-term cost model using a standardized 100 patient model.

Results:

Treatment with CDT resulted in fewer bleeding complications (4.2 percent vs. 13.8 percent, p=0.005), a shorter length of stay (3.44 days vs. 6.47 days, p<0.001), a greater percentage of patients returning to their prior living conditions (89.0 percent vs. 79.3 percent, p=0.042), and a lower rate of chronic pulmonary hypertension at 12 months (6.3 percent vs. 15.9 percent, p=0.030) than those treated with systemic heparin. The expense of treatment utilizing CDT was greater than those undergoing systemic heparin treatment with a difference of approximately $31,000 (p=0.001) though our cost model showed the heparin group to have a higher cost over time.

Conclusion:

For patients with massive or submassive PE, this study demonstrated a significant long-term cost savings and improved outcomes for patients treated with catheter directed thrombolysis when compared to systemic heparin administration.

Introduction

Pulmonary embolism (PE) is a significant disease process that affects an estimated 117 cases per 100,000 person-years.¹ Mortality due to PE has traditionally been reported as significant though rates reported are variable depending on clinical presentation.^2,3 Mortality rates have been trending down since 2001.⁴ A recent study showed advanced therapies for treating PE was independently associated with decreased mortality.⁵ Unfortunately, survivors of the acute event are at risk for long-term complications as a result of this disease process. Complications and disease sequelae develop in a subset of this patient population with the most serious of these complications being chronic pulmonary hypertension (CPH). Further, a subset of patients will develop right sided heart failure.⁶ There is a significant cost in treating CPH which can range from $98,000–117,000 per year.⁷

There are three categories or classifications of pulmonary embolisms: minimal, submassive, and massive. Patients categorized as minimal have a low blood clot burden in the distal vascular segments of the lung. These patients often present with mild symptoms or asymptomatic and are treated with systemic anticoagulation. Patients in this category tend to have fewer complications. Submassive and massive PEs have a greater clot burden or clot(s) in the proximal pulmonary arteries. Patients categorized as having a submassive PE present with right ventricular dysfunction without systemic hypotension. Patients categorized as having a massive PE present with systemic hypotension, cardiac arrest, and/or cardiogenic shock. Outcomes for patients presenting with submassive and massive PE are worse than those categorized as minimal.^2,8 These patients are often treated with inpatient administered anticoagulation, which is the standard of care, or other aggressive techniques.^9,10 The true percentage of patients who develop CPH following a PE is unknown. Analysis has suggested the rate of CPH development to be as high as 43.5 percent based on previous studies.¹⁰

Current treatment modalities for submassive and massive PE include systemic heparin administration, systemic thrombolysis administration,¹¹ catheter directed thrombolysis,^8,12,13 catheter directed thrombectomy, percutaneous thrombectomy, and open surgical embolectomy.¹⁴ Literature supports the efficacy of these treatment modalities depending on patient characteristics, provider experience and hospital resources.^5,8,11 Current evidence for catheter-based interventions for PE are limited because they are primarily based on single-center studies.⁸ This has obvious limitations for guiding clinical decision making. Catheter-directed therapy is generally accepted for treatment of submassive or massive PE within current guidelines though there are no widely accepted standards on the amount or duration of lytic therapy.⁸ Even with the limitations, catheter-directed thrombolysis has shown significant reduction of right ventricle dilation in small scale randomized, controlled trials.¹⁵

There is a significant cost burden for society associated with the development of CPH though there are limited studies evaluating the difference in CPH rates between systemic heparin and CDT. This study evaluates catheter-directed thrombolysis compared to systemic heparin administration regarding patient outcomes, the rate of CPH development, short-term expense, and long-term expense associated with these two treatment modalities in a single center. A major Midwest center has been utilizing both catheter-directed thrombolysis and systemic heparin therapy in the treatment of submassive and massive PE with catheter-directed therapy since 2006.¹³

Methods

Data was collected as a retrospective chart review from patients treated from January 2011–November 2018. Patients were identified and abstracted from the electronic medical record (EMR) using the primary diagnosis codes for pulmonary embolism (Table 1). The Sanford Health Institutional Review Board (IRB) approved access to the patients’ health information for this retrospective study. Because the health information was to be de-identified, the IRB did not require the patients to undergo informed consent. The inclusion criteria for this study was treatment for pulmonary embolism. The exclusion criteria for this study was pre-existing pulmonary hypertension or the use of an alternate treatment modality than systemic heparin and CDT. Patient demographics, PE type, cost of admission and readmission if within thirty days, bleeding complications, discharge location, and the development of chronic pulmonary hypertension within twelve months of developing the PE were recorded. CTD technique and monitoring were followed as described by Nykamp, et al.¹³ Patient discharge location and status was recorded as documented in the patients initial discharge disposition.

Table 1.

ICD codes.

Code	ICO description
ICD-9
415.11	Iatrogenic pulmonary embolism and infarction
415.12	Septic pulmonary embolism
415.13	Saddle embolus of pulmonary artery
415.19	Other pulmonary embolism and infarction
ICD-10
126.01	Septic pulmonary embolism with acute cor pulmonale
126.02	Saddle embolus of pulmonary artery with acute cor pulmonale
126.09	Other pulmonary embolism with acute cor pulmonale
126.90	Septic pulmonary embolism without acute cor pulmonale
126.92	Saddle embolus of pulmonary artery without acute cor pulmonale
126.99	Other pulmonary embolism without acute cor pulmonale
127.82	Chronic pulmonary embolism

Baseline characteristics were reported as documented in the EMR at the time of admission. The PE location was determined via computed tomography angiography or ventilation-perfusion scans. Right heart strain was determined as being present if evidenced by right ventricular dilation, hypokinesis, or septal wall deviation as demonstrated by echocardiography or computed tomography. Bleeding complications were collected from the EMR during the initial admission and/or within 30 days of initial presentation. Cost was collected from the EMR via the hospital account for the associated admission and 30-day readmission, if applicable. Total costs were determined by the sum of the hospital and professional billing for each encounter with the average cost for each treatment modality calculated. Differences between the CDT and systemic heparin groups were compared using Welch’s t-test for continuous variables and chi-squared test for categorical variables. Significance was determined with p-values less than 0.05.

An additional cost model was completed based on the results of the short-term cost data using the average total cost to treat one patient and applying it over a population of 100 people to look at the cumulative cost to treat a population initially, at one year, and at five years. We were not able to collect data on long-term treatment due to out of system care following an event which includes nursing home, swing-bed, or additional hospitalizations. This additional model was based on 100 people to estimate and evaluate the potential cost burden to society. The cost difference between the treatment modalities was determined by averaging the total hospital and professional bills for each treatment group. The cost to treat and the rate of CPH from the data were included in the calculation to compare the treatment if 100 patients were treated. The cost per year to treat CPH was determined by multiplying the rate of CPH development for the respective groups and the estimated cost to treat an individual for one year. Based on claims from both commercial and Medicare Advantage enrollees, $100,000 was used as the estimated cost per year associated with the management of CPH.⁷ The cost to treat 100 patients for five years was determined by taking the one-year value multiplied by five. Finally, the five-year difference in total cost was the sum of the five-year cost to treat CPH and the initial cost difference in the treatment modalities.

Results

Between January 2011 and November 2018, 385 patients were identified using the described primary diagnosis codes. Twenty-four patients were excluded because they were treated with an alternative treatment modality (placement of IVC filter only or other anticoagulation method) and 20 were excluded due to having pre-existing CPH. The final study group included 341 patients with 283 patients treated with CDT and 58 patients treated with systemic heparin administration. Table 2 lists the baseline characteristics for the respective cohorts. There was a 10-year difference in the average age between the two groups with an average age of 68.5 and 58.8 for the heparin and CDT groups respectively.

Table 2.

Baseline characteristics.

Baseline characteristics	CDT (n=283)	Systemic heparin (n=58)	p-value
Age (years)	58.83	68.47	<0.001
Male	151 (53%)	31 (53%)	0.990
Female	132(47%)	27(47%)
Preexisting conditions
Chronic obstructive pulmonary disease	22(8%)	11 (19%)	0.009
Hypertension	131 (46%)	28 (48%)	0.782
Hypercholesterolemia	101 (37%)	17(29%)	0352
Recent trauma	11(4%)	1(2%)	0.415
Recent major surgery	34(12%)	12(21%)	0.078
Diabetes	51 (18%)	2(3%)	0.005
Renal insufficiency (creatinine >15 mg/dL)	8(3%)	4(7%)	0.125
Coronary artery disease	29(10%)	13 (22%)	0.010
Peripheral vascular disease	6(2%)	3(5%)	0.186
Current or recent cancer	39 (14%)	23(40%)	<0.001
Identified hypercoagulable disorder	12(4%)	4(7%)	0383
Previous deep vein thrombosis	43(15%)	13(22%)	0.176
Previous pulmonary embolism	34 (12%)	7(12%)	0.991
Idiopathic	62(22%)	7(12%)	0.089

Values listed are n-values with percent of occurrence lasted in parentheses.

There was a significantly higher number of patients treated with CDT who presented with right heart strain than those treated with systemic heparin (58.0 percent vs. 22.4 percent, p <0.001) though there was no difference in patients presenting with hypotension. Patient outcomes were significantly better for those treated with CDT with a reduced length of stay (3.4 days vs. 6.5 days, p<0.001), a lower rate of pulmonary hypertension at twelve months (6.3 percent vs. 15.9 percent, p=0.030), a lower rate of readmission within 30 days (6.4 percent vs. 15.5 percent, p=0.019), a greater percentage of patients returning to their prior living conditions (89.0 percent vs. 79.3 percent, p=0.042), and fewer bleeding complications (4.2 percent vs. 13.8 percent, p=0.005). The mortality rate for the CDT group was almost half that for the heparin group (3.9 percent vs. 6.9 percent) but this did not meet statistical significance (Table 3). These differences may be explained by the 10-year age difference between the two treatment groups.

Table 3.

Results of data analysis.

Variable	CDT n=283	Systemic heparin n=58	p-value
Length of stay (days)	3.44	6.47	<0.001
Pulmonary hypertension by 12 months	14 (6.3)	7(15.9)	0.030
Right heart strain	164 (58.0)	13 (22.4)	<0.001
Hypotension	36 (12.7)	6 (103)	0.616
Thirty-day mortality	11 (3.9)	4(6.9)	0.309
Thirty-day readmission	18 (6.4)	9 (155)	0.019
Return to phot living conditions	262 (89.0)	46(79.3)	0.042
Bleeding complications	12(4.2)	8 (13.8)	0.005

Values listed are n-values with percent of occurrence listed in parentheses

The cost associated with the initial admission was significantly higher in the CDT group when compared to heparin (p=0.001) with a difference of $31,000. However, the cost analysis shows the heparin group to have a higher long-term cost to treat.

Cost analysis of the initial cost to treat 100 patients showed an approximate difference of $3,100,000 more for the CDT group. The cost to treat patients who may go on to develop CPH in the CDT group was estimated as $630,000 at one year and $3,150,000 at five years. The systemic heparin group had an estimated cost of $1,590,000 and $7,950,000 for the patients who may go on to develop CPH at one year and five years, respectively. The total difference in cost to treat was $6,250,000 for the CDT group and $7,950,000 for the systemic heparin treated group (Table 4). This resulted in a difference of $1.7 million cost savings in the CDT group when the total cost to treat was evaluated out to five years.

Table 4.

Standardized cost calculation treating 100 patients with pe and resulting CPH.

Value description	CDT		Heparin
Patients		100	100
Approximate difference in the cost to treat for initial hospitalization per patient		$31k	-
Difference in the initial cost to treat (hospitalization)	$3.1 M	-
Rate of pulmonary HTN		63%	15.9%
Cost per year to treat CPH		$100k	$100k
Cost to treat 100 patients with CPH per year		$630k	$1.59M
Cost to treat 100 patients with CPH over 5 years	$3.15M	$7.95M
Difference in the 5-year total cost to treat	$6.25M	$7.95M

(Costs recorded are USD, k=1,000 USD and M=1,000,000 USD)

Discussion

The difference in short-term costs were significant between the two groups due to those patients treated with systemic heparin administration having a lower cost to treat upon admission. The difference is due to the cost incurred for use of a catheter device for the CDT group. There was an expense associated with both placing the device and the cost of the catheter system itself (list price of $500–3,000 per catheter), as well as the significant cost of the lytic drug. Systemic heparin utilizes both a cost effective IV system and an inexpensive drug. While the short-term cost of CDT was higher, the long-term costs were reduced due to a lower percentage of patients developing chronic pulmonary hypertension. Based on the analysis, the difference in cost savings over a five-year period was estimated at $1.7 million in the CDT group compared to the group treated with systemic heparin. This suggests a significant reduction in the long-term cost burden for society. Further evaluation of the data indicates advantages of CDT for patients which can include a lower incidence in readmission, reduction in bleeding complications, 50 percent reduction in length of stay, and a higher rate of returning to prior living conditions. These findings suggest improved quality of life outcomes for patients treated with CDT vs. systemic heparin administration though the cost associated with these factors is difficult to analyze. Conceptually, patients who recover faster can return to work earlier which reduces the economic burden for the patient and healthcare payers. A significantly higher number of patients returned to their prior living conditions in the group treated with CDT which would decrease the cost associated with skilled nursing facilities, swing beds and other care facilities. While the costs of these care facilities were not specifically addressed in this study, there is a substantial expense that should be considered in association with placement in a care facility. If this cost were to be factored in, the savings over five years for those treated with CDT vs systemic heparin administration would likely increase.

In this model, when the potential cost burden to society was standardized across 100 patients, we show an initial higher cost to treat with catheter directed therapy. However, the data also demonstrated a long-term return on investment due to decreased complication rate and reduction in the cost of long-term care. The ability for patients to return to their prior living conditions without pulmonary hypertension is important in this data set as our patients had an average age of 58.8 years for the catheter directed group and on average will live 10 years longer than the systemic heparin group that we studied. The analysis shows an estimated savings over five years when implementing the catheter directed therapy and treating 100 patients. The calculation demonstrates that there is a 1.7-year cost to recover. Even with the increased cost to treat initially, the authors believe this therapy has promise to show a cost savings in the population health setting. This also suggests that we should find ways to reduce the cost to treat with catheter-directed therapy. These include reducing the cost of the device, procedural costs, and placement of accessory devices such as IVC filter. This would further reduce the initial cost to treat and result in a greater cost savings long-term for those treated with CDT.

The data gathered from this study also suggests that the utilization of CDT for patients who are more clinically stable, but still fall into the submassive category, should be further studied. The study showed that despite the more critical presentation in the CDT group there was a lower percentage in the development of CPH. This raises the question of whether patients presenting with a submassive PE, i.e., normotensive without right heart strain but with a significant clot burden should be treated with CDT to reduce the risk of developing CPH. Our CPH rates were higher than historical rates of 1–4 percent. The results are within reasonable variance as the true rates are reported to be unknown and estimated up to 43.5 percent which is actually closer to what we see.¹⁰

Limitations

This study has some potential limitations. First, the ICD 10 codes used were to determine if a patient had been treated for pulmonary embolism, not to determine if there was a diagnosis of pulmonary hypertension. There was also a discrepancy between the numbers of patients treated with each modality. This could have affected the data used in the long-term cost model. Additionally, the model was not a direct measurement of long-term cost. While this could affect the results reported, the model does give an indication of cost impact between these two methods of treatment specifically related to the development of CPH.

Conclusion

The data from this study suggests patients treated with CDT for submassive and massive PEs have improved outcomes. As we move toward population health, new therapies need to be value based and should provide a cost savings over the standard of care while also providing a better quality of life. The latter does not have a direct monetary value; however, this is an important factor to society. Some consideration should be made to whether all submassive and massive pulmonary embolisms should be treated with catheter-directed thrombolysis.