A Comparison of Extremity Thrombosis Rates in Adolescent and Young Adult Versus Younger Pediatric Oncology Patients at a Children's Hospital

Abstract

Purpose:

To examine whether the rates of thrombosis in children (≤14 years of age) and adolescent/young adult (AYA) patients (15–22 years of age) with cancer is different.

Methods:

We retrospectively studied the rates of thrombosis in children and AYA patients at the Children's Hospital of Pittsburgh during the years 2002–2010, using the tumor registry database. This list was then divided into two groups based on age at diagnosis. A review of ICD-9 codes from hospital billing records was then performed to identify patients who carried diagnoses of cancer (140.x–239.x) and venous thrombosis of the extremities/vena cavae (453.x) simultaneously. This list was confirmed by electronic medical record review. Proportions, comparisons, and descriptive statistics were then performed.

Results:

One thousand three hundred nine total patients were identified; 274 patients fit into the AYA age category (mean age 17.3 years) and 1036 patients were in the child group (mean age 6.5 years). Overall, 30 patients (2.29%) had thrombosis: 4.76% of the AYA patients (13/273) and 1.64% of the child group (17/1036). The difference in these proportions had a p-value = 0.004.

Conclusions:

This study suggests that the risk of extremity deep vein thrombosis is higher in the AYA subset of oncology patients than in the patients who are 14 years or younger. Prospective studies to elucidate the true rate of thrombosis, as well as to study the benefit of prophylactic anticoagulation in the AYA population, should be undertaken.

Introduction

Cancer is a well-known risk factor for the development of venous thromboembolism (VTE) in adults.^1–7 Traditionally, as the risk of VTE is higher in adults than it is in children, this connection has not been investigated as frequently in pediatric oncology. Increasing evidence in pediatric oncology suggests that the risk of thrombosis is lower but still an important concern.^8–10

While young patients with cancer may not have the same risk of thrombosis as do older patients, it is an increasingly recognized complication. Previous studies that have evaluated specific high-risk populations, such as acute lymphoblastic leukemia (ALL) patients receiving asparaginase therapy have found variable amounts of VTE and sporadically useful prophylactic interventions, though often the number of patients was too low to provide statistical significance.^11–14

In recent years, studies in pediatric hematology have shown that the occurrence of thrombosis is higher than previously thought, especially in hospitalized children.^15–17 Severe and/or chronic illnesses, implanted medical devices, inherited thrombophilic traits, and improved intensive care practices are some of the proposed reasons for increased detection and awareness of VTE in children. Central venous catheters and access devices are the most common risk factor identified in pediatric VTE, though they are often required for the care of illnesses that may also provoke VTE.¹⁸

Adolescent and young adult (AYA) oncology patients have unique physical and psychosocial characteristics that pose new challenges in oncologic management.^19–21 Some of these factors are accounted for by the unique social and developmental changes that young adults face compared to children and older adults. Some studies of AYA oncology patients also suggest that the behavior of their tumors is different than that in younger and older patients.²¹ It now also is known that VTE in childhood has bimodal peaks, one of which is adolescence (the other being infancy).^22,23 Infants are relatively hypercoagulable, due to delayed development of natural anticoagulants. Adolescents, on the other hand, may have more comorbidities, higher rates of trauma, and the emergence of age-related vascular and inflammatory changes. This increase could mean that AYA oncology patients may have a higher VTE rate than younger patients.

Though the AYA age group is defined as 15–39 years of age, the care of AYA patients often is split between pediatric and adult centers depending on disease type and patient age. Similarly, the care of young adults with thromboses can take place across the same two settings depending on age, local expertise, and procedural requirements. This can contribute to difficulties in comparing studies of VTE in patients with cancer of different ages.

This study sought to address this question by examining whether the rates of thrombosis in children and AYA patients with cancer are different and if any characteristics of these two groups were predictive of VTE risks.

Methods

This study was approved by the University of Pittsburgh Institutional Review Board as an exempt protocol. We retrospectively analyzed the tumor registry database at the Children's Hospital of Pittsburgh for the years 2002–2010. A list of all oncology patients evaluated at our center was generated and then divided into the two age group categories to provide the total number of subjects in our population. Using the general standards of our institution, we established the age range of pediatric patient as less than or equal to 14 years of age and of AYA patients as 15–22 years of age.

Simultaneously, a review of ICD-9 codes from hospital billing records was then performed to identify patients who carried diagnoses of cancer (140.x–239.x) and venous thrombosis (453.x). The ICD-9 codes of 453.x include VTE of the inferior vena cava and acute and chronic VTE of upper and lower extremities. Patients with superficial venous thromboses only were excluded, as were patients with only cerebral sinus venous thrombosis (CSVT), pulmonary embolism (PE), or portal vein thrombosis (PVT). This list of patients with thrombosis and cancer was then used to perform an electronic medical record review. Age and weight were recorded as numerical variables, while AYA status and gender were recorded as dichotomous variables.

Type of cancer diagnosis was classified as select categorical variables: ALL, acute myeloid leukemia, Hodgkin lymphoma, non-Hodgkin lymphoma, central nervous system (CNS) tumors, osteosarcoma+Ewing sarcoma, rhabdomyosarcoma, nonrhabdomyosarcoma soft-tissue sarcomas, and other cancers (primarily Wilms tumor, hepatoblastoma, and neuroblastoma). Central-venous catheter association and the presence of a thrombophilic trait also were categorized as dichotomous variables. Thrombophilia testing was defined as levels of protein C, protein S, antithrombin III, and factor VIII along with direct testing for factor V mutations, the prothrombin gene variant, and lupus anticoagulant or antiphospholipid antibody panels.²²

Proportions, comparisons, and descriptive statistics were then performed from the resulting data set. A Fisher's exact test was used to compare the proportion of patients with thrombosis in each age group. Logistic regression was used to compute odds ratios. A p-value of 0.05 was used as the threshold for statistical significance. Stata/SE 13.1 (StataCorp, College Station, TX) was used to perform all statistical analyses.

Results

A total of 1309 oncology patients were identified from the tumor registry; 273 patients fit into the AYA age category (mean age 17.3 years) and 1036 patients were in the child group (mean age 6.5 years). Medical record review yielded 30 patients with cancer and a thrombosis (2.29%). Combining these data sets showed that 4.76% of the AYA patients (13/273) and 1.64% of the child group (17/1036) had a thrombosis diagnosed at presentation of their cancer or during their oncology treatment.

Table 1 displays the overall results. The Fisher's exact test showed that AYA group had a statistically significant higher proportion of patients with thrombosis, with a p-value = 0.004. The Fisher's exact test was used in place of a chi-squared test due to the low absolute numbers in the thrombosis groups. Logistic regression was performed to determine odds ratios. The odds of AYA patients having a thrombosis were 2.98 times higher than the same odds of younger patients. This featured a 95% confidence interval of 1.43–6.22.

Table 1.

Comparison of Thrombosis Rates

Age group	Total number of patients	Number of patients with VTE	Proportion, %
Child	1036	17	1.6
AYA	273	13	4.76

p = 0.004.

AYA, adolescent and young adult; VTE, venous thromboembolism.

Additionally, the odds ratio of AYA patients having a thrombophilic trait compared to younger patients was 1.94, though thrombophilia was tested in only a small portion of patients overall. Not all patients had identical thrombophilia testing performed; consequently, this result did not meet statistical significance, as the 95% confidence interval was 0.745–5.06. Only 14 of the patients who developed thromboses underwent thrombophilia testing (9 child and 5 AYA). Perhaps the most common acquired thrombophilic state in children, the presence of a central venous catheter, did not have a statistically significant role between the two groups. Six patients (three in each group) had central venous line (CVL)-associated thromboses. These patients, somewhat surprisingly, represented a minority of all patients with thromboses.

No other statistically significant associations were found in the analysis. This included no statistical significance to type of cancer diagnosis or gender. Table 2 shows the overall breakdown of cancers and the distribution within each age group category. The highest number of patients with a thrombosis was children with CNS tumors, while the highest proportion of patients with thromboses was AYA patients with Hodgkin lymphoma. Table 3 details select characteristics of patients who developed thromboses. Family history of thromboses was not well documented in the medical record and was not included in the analysis.

Table 2.

Overview of Thromboses by Cancer Type and Age Group

	Total patients	Patients with thrombosis	%	Children with cancer	Children with cancer and thrombosis	%	AYA patients with cancer	AYA cancer and thrombosis	%
ALL	260	6	2.31	231	5	2.16	29	1	3.45
AML	93	3	3.23	68	2	2.94	25	1	4.00
Non-Hodgkin lymphoma	66	4	6.06	47	2	4.26	19	2	10.53
Hodgkin lymphoma	91	3	3.30	48	0	0.00	43	3	6.98
CNS tumors	341	7	2.05	288	6	2.08	53	1	1.89
Osteosarcoma+Ewing sarcoma	158	3	1.90	115	1	0.87	43	2	4.65
Rhabdomyosarcoma	77	2	2.60	56	1	1.79	21	1	4.76
Nonrhabdo soft-tissue sarcoma	38	0	0.00	34	0	0.00	4	0	0.00
Other cancers	185	2	1.08	149	0	0.00	36	2	5.56
	1309	30	2.29	1036	17	1.64	273	13	4.76

ALL, acute lymphoblastic leukemia; AML, acute myeloid leukemia; CNS, central nervous system.

Table 3.

Selected Details About Patients Who Developed Thromboses

Patient age (years)	Oncologic diagnosis	Was VTE in same vessel as CVL?	Thrombophilia testing results
0.3	CNS tumor	No	n/a
0.6	CNS tumor	Yes	n/a
0.8	Osteosarcoma	Yes	Negative
1.0	CNS tumor	Yes	n/a
1.3	Rhabdomyosarcoma	No	Heterozygous prothrombin G20210A mutation; elevated factor VIII level
1.3	ALL	No	n/a
2.3	Non-Hodgkin lymphoma	No	Negative
3.2	CNS tumor	No	Negative
3.8	CNS tumor	No	Negative
4.3	ALL	No	Heterozygous factor V Leiden mutation; heterozygous protein S deficiency
4.6	AML	No	Negative
6.9	Non-Hodgkin lymphoma	No	n/a
7.6	ALL	No	n/a
11.8	CNS tumor	No	n/a
13.3	ALL	No	Negative
13.7	AML	No	Heterozygous protein C deficiency
14.5	ALL	No	n/a
15.2	Non-Hodgkin lymphoma	No	n/a
15.9	Hodgkin lymphoma	Yes	Lupus anticoagulant
15.9	Other	No	Lupus anticoagulant+Anticardiolipin antibodies
16.0	Ewing sarcoma	No	n/a
16.1	Hodgkin lymphoma	No	n/a
16.1	Rhabdomyosarcoma	No	n/a
16.3	Hodgkin lymphoma	No	n/a
16.7	AML	No	Heterozygous factor V Leiden mutation
17.2	Non-Hodgkin lymphoma	Yes	Elevated factor VIII level
17.5	Other	No	n/a
17.9	CNS tumor	No	Negative
20.7	Ewing sarcoma	No	n/a
21.9	ALL	Yes	n/a

CVL, central venous line.

Discussion

This study compared the rates of thrombosis in adolescent and young adult oncology patients and children 14 years or younger. Our results suggest that the risk of deep vein thrombosis (DVT) is higher in AYA patients than in younger children. Through isolating the different thrombosis rates in two categories of pediatric patients, this study adds to the evidence that AYA oncology patients are a unique population.

Though our study found lower overall rates of thrombosis than others have done in the past,^24,25 we were able to confirm that the risk of VTE is higher in the AYA population. This risk appeared to be independent of cancer diagnosis, gender, central venous catheter placement, and thrombophilic traits. The presence or absence of concomitant medications, especially hormonal agents, was reviewed but did not allow for a sufficiently powered analysis. Estrogen-containing medications (the primary VTE risk from contraceptive and menstrual-regulating medications) are often avoided in patients with inflammatory diseases. Other types of hormonal agents have low or no risk of DVT and were deemed to not be a likely confounder in the analysis. Finally, weight was evaluated but not felt to be a reliable variable given the differences in growth curves and BMI values for toddlers, young children, and adolescents.

Thrombosis and cancer have been linked for a number of years, though their precise mechanism of connection remains under active study.^1,26–29 Traditionally, as the risk of VTE is higher in adults than it is in children, investigations and treatments focusing on VTE and cancer predominantly have been carried out in adult patients. The rates of VTE in children have been increasing in recent time, however, improved care of systemic illnesses are one of the proposed links.^16,17,30 Identifying specific at-risk populations within pediatrics will allow for targeted trials for prophylaxis and therapeutic interventions.^5,9,28,31 In this study, the increased rate of thrombosis found in AYA oncology patients parallels the increased rate of VTE in nononcology AYA patients. Given that, it is possible that establishing the benefits of anticoagulation prophylaxis in oncology patients would best be served in the higher risk age groups before applying it to all young people.

Our study's limitations relate primarily to its retrospective design, its single-institution data source, and its narrowed focus on peripheral extremity VTE. Retrospective studies that are based on medical records face a number of difficulties. Especially when billing records and coding are used, errors of omission may be common, possibly altering the true number of events analyzed. Retrospective studies do not have the ability to detect medical record documentation errors.

On a related level, though our institution did have a considerable overall number of patients, differences with published national cancer rates can be seen. In our sample, there was a lower percentage of patients with leukemia and higher percentages of patients with central nervous-system tumors and sarcomas than what was published in the first few years of the Children's Oncology Group's new registry.³² This variance may be due to temporal factors of diagnosis patterns or to institutional differences—for example, the availability of phase I or other clinical trials not present at other centers. Our study also capped the AYA aged-group at 22 years. Though this reduced the number of possible AYA patients with cancer, our sample appeared to have fewer AYA patients than expected, using the National Cancer Institute's Surveillance, Epidemiology, and End Results incidence data would suggest.³³ This again may be due to local referral patterns; nevertheless, it is reasonable to believe that a larger sample of AYA patients would reinforce the overall conclusion of a higher DVT rate, given the statistical significance seen with our sample.

Finally, our study focused on peripheral VTE, excluding CSVT, PE, and PVT most notably. The treatment choices posed by those two types of thromboses may pose different challenges (CSVT with hemorrhagic conversion) or in etiology (PVT) than treatment of peripheral VTE.^34,35 This focus of our project may account for the lower overall rate of thromboses in our oncology patients than what has been reported previously in studies from other centers.

Despite these limitations, we believe this study adds evidence to the important and growing distinction between AYA and younger oncology patients. As survival rates continue to improve in oncology, studies such as ours that address quality-of-life and supportive care concerns will be of added benefit in improving the overall care of our patients.

Prospective studies to elucidate the true rate of thrombosis in the AYA and pediatric populations should be undertaken. Guidelines for treatment of VTE in the setting of cancer currently do not differ from the treatment of VTE in general settings,³⁴ though studies of prophylactic regimens and of new anticoagulant agents are emerging. The importance of cooperative groups, such as the Children's Oncology Group, cannot be understated in helping pool resources to study these diseases more completely. Furthermore, our study supports the belief that AYA-specific cooperative group clinical trials in oncology are warranted and needed.

Footnotes

Acknowledgments

Dr. J.D.C. performed primary data collection though all authors participated in data analysis and interpretation. All authors have reviewed this article, approve its submission, and take full responsibility for its completion. Dr. J.D.C. wrote the first draft of the article but all authors were actively involved in revisions. Preliminary results from this study were presented at the American Society of Pediatric Hematology/Oncology Annual Meeting in Baltimore, Maryland, May 2011.

Author Disclosure Statement

The authors have no relevant conflicts of interest to declare. No honoraria or sponsors were involved in the creation of this article.

References

Connolly

, Khorana

. Risk stratification for cancer-associated venous thromboembolism. Best Pract Res Clin Haematol. 2009; 22(1):35–47.

Falanga

, Marchetti

. Venous thromboembolism in the hematologic malignancies. J Clin Oncol. 2009; 27(29):4848–57.

Kakkar

. Prevention of venous thromboembolism in the cancer surgical patient. J Clin Oncol. 2009; 27(29):4881–4.

Lyman

, Khorana

. Cancer, clots and consensus: new understanding of an old problem. J Clin Oncol. 2009; 27(29):4821–6.

Rana

, Levine

. Prevention of thrombosis in ambulatory patients with cancer. J Clin Oncol. 2009; 27(29):4885–8.

Sousou

, Khorana

. Identifying cancer patients at risk for venous thromboembolism. Hamostaseologie. 2009; 29(1):121–4.

Timp

, et al. Epidemiology of cancer-associated venous thrombosis. Blood. 2013; 122(10):1712–23.

Bajzar

, et al. Thrombosis in children with malignancy. Curr Opin Pediatr. 2006; 18(1):1–9.

Lee

, Peterson

. Treatment of cancer-associated thrombosis. Blood. 2013; 122(14):2310–7.

10.

Walker

, et al. Venous thromboembolism in children with cancer—a population-based cohort study. Thromb Res. 2014; 133(3):340–4.

11.

Abbott

, et al. The impact of prophylactic fresh-frozen plasma and cryoprecipitate on the incidence of central nervous system thrombosis and hemorrhage in children with acute lymphoblastic leukemia receiving asparaginase. Blood. 2009; 114(25):5146–51.

12.

Harlev

, et al. Prophylactic therapy with enoxaparin in children with acute lymphoblastic leukemia and inherited thrombophilia during L-asparaginase treatment. Thromb Res. 2010; 126(2):93–7.

13.

Meister

, et al. Comparison of low-molecular-weight heparin and antithrombin versus antithrombin alone for the prevention of symptomatic venous thromboembolism in children with acute lymphoblastic leukemia. Pediatr Blood Cancer. 2008; 50(2):298–303.

14.

Mitchell

, et al. Trend to efficacy and safety using antithrombin concentrate in prevention of thrombosis in children receiving l-asparaginase for acute lymphoblastic leukemia. Results of the PAARKA study. Thromb Haemost. 2003; 90(2):235–44.

15.

O'Brien

, et al. Venous thromboembolism and adolescent and young adult oncology inpatients in US children's hospitals, 2001 to 2008. J Pediatr. 2011; 159(1):133–7.

16.

Streiff

. Diagnosis and initial treatment of venous thromboembolism in patients with cancer. J Clin Oncol. 2009; 27(29):4889–94.

17.

Takemoto

, et al. Hospital-associated venous thromboembolism in children: incidence and clinical characteristics. J Pediatr. 2014; 164(2):332–8.

18.

Journeycake

, Buchanan

. Catheter-related deep venous thrombosis and other catheter complications in children with cancer. J Clin Oncol. 2006; 24(28):4575–80.

19.

Albritton

, et al. Closing the Gap: Research and Care Imperatives for Adolescents and Young Adults with Cancer. National Cancer Institute LIVESTRONG Young Adult Alliance, 2006.

20.

Hayes-Lattin

, Mathews-Bradshaw

, Siegel

. Adolescent and young adult oncology training for health professionals: a position statement. J Clin Oncol. 2010; 28(32):4858–61.

21.

Hayes-Lattin

, et al. Closing the Gap: A Strategic Plan. National Cancer Institute LIVESTRONG Young Adult Alliance, 2006.

22.

Goldenberg

, Bernard

. Venous thromboembolism in children. Pediatr Clin North Am. 2008; 55(2):305–22, vii.

23.

Raffini

, et al. Dramatic increase in venous thromboembolism in children's hospitals in the United States from 2001 to 2007. Pediatrics. 2009; 124(4):1001–8.

24.

Grace

, et al. The frequency and management of asparaginase-related thrombosis in paediatric and adult patients with acute lymphoblastic leukaemia treated on Dana-Farber Cancer Institute consortium protocols. Br J Haematol. 2011; 152(4):452–9.

25.

Mitchell

, et al. A prospective cohort study determining the prevalence of thrombotic events in children with acute lymphoblastic leukemia and a central venous line who are treated with L-asparaginase: Results of the Prophylactic Antithrombin Replacement in Kids with Acute Lymphoblastic Leukemia Treated with Asparaginase (PARKAA) Study. Cancer. 2003; 97(2):508–16.

26.

Boccaccio

, Comoglio

. Genetic link between cancer and thrombosis. J Clin Oncol. 2009; 27(29):4827–33.

27.

Carrier

, et al. Management of challenging cases of patients with cancer-associated thrombosis including recurrent thrombosis and bleeding: guidance from the SSC of the ISTH. J Thromb Haemost. 2013; 11(9):1760–5.

28.

Levine

. New antithrombotic drugs: potential for use in oncology. J Clin Oncol. 2009; 27(29):4912–8.

29.

Magnus

, et al. Brain neoplasms and coagulation. Semin Thromb Hemost. 2013; 39(8):881–95.

30.

Setty

, O'Brien

, Kerlin

. Pediatric venous thromboembolism in the United States: a tertiary care complication of chronic diseases. Pediatr Blood Cancer, 2012; 59(2):258–64.

31.

Lee

. Anticoagulation in the treatment of established venous thromboembolism in patients with cancer. J Clin Oncol. 2009; 27(29):4895–901.

32.

Spector

, Ross

, Olshan

. Children's Oncology Group's 2013 blueprint for research: epidemiology. Pediatr Blood Cancer. 2013; 60:1059–62.

33.

Howlader

, Noone

, Krapcho

, Miller

, Bishop

, Altekruse

, Kosary

, Yu

, Ruhl

, Tatalovich

, Mariotto

, Lewis

, Chen

, Feuer

, Cronin

(eds). SEER Cancer Statistics Review, 1975-2013. Bethesda, MD: National Cancer Institute; based on November 2015 SEER data submission, posted to the SEER web site, April 2016. Accessed September 1, 2016 from: http://seer.cancer.gov/csr/1975_2013/

34.

Monagle

, et al. Antithrombotic therapy in neonates and children: antithrombotic therapy and prevention of thrombosis, 9th ed: American College of Chest Physiciancs Evidence-Based Clinical Practice Guidelines. Chest. 2012; 141(2):e737S–801S.

35.

Williams

, Chan

AKC

. Neonatal portal vein thrombosis: diagnosis and management. Semin Fetal Neonatal Med. 2011; 16(6):329–39.