Abstract
Rationale: Idiopathic interstitial pneumonia is characterized by pulmonary fibrosis and high mortality.
Objectives: We examined the association between ever-diagnosed venous thromboembolism and risk of incident idiopathic interstitial pneumonia. Venous thromboembolism was taken as a proxy for a procoagulant state in an individual.
Methods: We conducted a study of the entire Danish population from 1980 through 2007, comprising 7.4 million individuals. Incident idiopathic interstitial pneumonia, ever-diagnosed venous thromboembolism, and use of prescription anticoagulants were drawn from national Danish registries.
Measurements and Main Results: Age-standardized incidence rates per 10,000 person-years for idiopathic interstitial pneumonia were higher among those ever diagnosed with venous thromboembolism (1.8; n = 158,676), pulmonary embolism (2.8; n = 70,586), and deep venous thrombosis only (1.2; n = 88,090), than among control subjects (0.8; n = 7,260,278). Multivariate-adjusted hazard ratios for idiopathic interstitial pneumonia were 1.8 (95% confidence interval [CI], 1.7–1.9) in those ever diagnosed with venous thromboembolism, 2.4 (95% CI, 2.3–2.6) in those ever diagnosed with pulmonary embolism, and 1.3 (95% CI, 1.2–1.4) in those ever diagnosed with deep venous thrombosis only, compared with control subjects. Corresponding hazard ratios in those ever diagnosed with venous thromboembolism stratified in those ever and never treated with anticoagulants were 1.4 (95% CI, 1.2–1.6) and 2.8 (95% CI, 2.4–3.1) (venous thromboembolism × anticoagulation use interaction on idiopathic interstitial pneumonia outcome: P = 1.5 × 10−10).
Conclusions: In the general population, ever-diagnosed venous thromboembolism was associated with idiopathic interstitial pneumonia, particularly among those never treated with anticoagulants.
There are no established risk factors for idiopathic interstitial pneumonia. This disease has a median survival from diagnosis of only 3 years. We hypothesized that individuals in a procoagulant state are at increased risk of idiopathic interstitial pneumonia, possibly through lifelong continuous development of unnoticed small pulmonary emboli leading to progressive pulmonary fibrosis and eventually idiopathic interstitial pneumonia.
We examined the entire Danish population from 1980 through 2007, including 7.4 million individuals. We showed that in the general population ever-diagnosed venous thromboembolism was associated with idiopathic interstitial pneumonia, particularly among those never given anticoagulant treatment
We previously showed that homozygotes for the coagulation factor V R506Q (i.e., factor V Leiden) polymorphism have an 18-fold increased risk of venous thromboembolism (2) and suffer from severe dyspnea, reduced lung function, and increased lung function decline (3). We therefore speculated that individuals homozygous for factor V R506Q lifelong continuously develop clinically unnoticed small pulmonary emboli and increased lung fibrosis. Following this, it is also possible that many individuals with ever-diagnosed venous thromboembolism throughout most of their life are in a procoagulant state, and may develop progressive lung fibrosis through clinically unnoticed small pulmonary emboli, and eventually idiopathic interstitial pneumonia and interstitial lung disease.
We examined the association between venous thromboembolism (comprising pulmonary embolism and deep venous thrombosis) and risk of idiopathic interstitial pneumonia and interstitial lung disease. Thus, we used a diagnosis of venous thromboembolism as a proxy for individuals in a procoagulant state, and examined whether such individuals have a lifetime increased risk of idiopathic interstitial pneumonia and interstitial lung disease. This is an important question to study because at the moment there is no established treatment for idiopathic interstitial pneumonia and the median survival from diagnosis is only 3 years (1).
For this purpose we studied the entire Danish population from 1980 through 2007 and used information from the national Danish Patient Registry, the national Danish Causes of Death Registry, the national Danish Civil Registration System, and Statistics Denmark; these four registries were all complete from 1980 through 2007. Furthermore, using information from the national Danish Registry of Medicinal Products Statistics from 1995 through 2007, we compared risk of idiopathic interstitial pneumonia and interstitial lung disease in those ever diagnosed with venous thromboembolism, who ever versus never had been treated with anticoagulants in the form of vitamin K antagonists. Some of these results have been previously reported in the form of an abstract (4).
We conducted a study of the entire Danish population from 1980 through 2007, comprising 7,418,953 individuals. The national Danish Civil Registration System records all births, deaths, emigrations, and immigrations in Denmark, recorded by the civil registration number, which is unique to every person living in Denmark and includes information about age and sex (5). These studies were approved by Herlev Hospital, Copenhagen University Hospital, and Statistics Denmark; in Denmark, approval by ethics committees of such anonymous nationwide studies is not necessary.
Information about interstitial lung diseases (idiopathic interstitial pneumonia and other interstitial lung diseases) was drawn from the national Danish Patient Registry (85% of events) and the national Danish Causes of Death Registry (15% of events).
The national Danish Patient Registry records information on discharge diagnoses from all Danish hospitals including outpatients, using the unique civil registration number. Records include admission date and diagnoses according to the World Health Organization (WHO, Geneva, Switzerland) International Classification of Diseases (ICD8 until 1993, thereafter ICD10). All individuals with records of discharge diagnoses for idiopathic interstitial pneumonia (ICD8 517; ICD10 J84.1) and interstitial lung disease (ICD8 515, 516, 517; ICD10 J84, J60-J70) from 1980 through 2007 were used in the study as incident end points.
The national Danish Causes of Death Registry records information about death date and causes of death for all deaths in Denmark, using the unique civil registration number, reported by hospitals and general practitioners. All individuals with records of causes of death diagnoses (using ICD8 and ICD10 codes as described previously) for idiopathic interstitial pneumonia and interstitial lung disease from 1980 through 2007 were also used in the study as incident end points.
Information about venous thromboembolism (pulmonary embolism and deep venous thrombosis) was drawn from the national Danish Patient Registry and the national Danish Causes of Death Registry, as described previously. All individuals with records of discharge diagnoses and/or causes of death diagnosis for deep venous thrombosis (ICD8 451, 671; ICD10 I80, O22.3, O87.1) and pulmonary embolism (ICD8 450, 673.99; ICD10 I26, O88.2) from 1980 through 2007 were used to identify individuals likely to be in a procoagulant state, and were used as the predictive variable in the study.
Statistics Denmark, covering all persons living in Denmark, records information about ethnicity, educational level, and geographical residency, using the unique civil registration number.
Information about the use of anticoagulants was obtained from the national Danish Registry of Medicinal Products Statistics, from 1995 through 2007. This registry records information about all prescribed drugs purchased in Danish pharmacies from 1995 and onward. Drugs administered during a hospital stay are not included, but immediately on discharge patients buy their own medications in Denmark. We obtained information about the use of anticoagulants, defined as vitamin K antagonists (ATC code B01AA).
Statistical analyses were performed with STATA 10.0 MP (StataCorp, College Station, TX) software. We assessed the association between ever-developed venous thromboembolism and incident idiopathic interstitial pneumonia and interstitial lung disease, by following all persons living in Denmark from 1980, from birth or time of immigration (whichever came last), to an idiopathic interstitial pneumonia or interstitial lung disease event, death, emigration, or end of 2007 (whichever came first); when individuals first emigrated and later immigrated back to Denmark, they were still included in the analysis.
Age-standardized incidence rates were calculated according to the WHO World Standard population (6). We used Kaplan-Meier curves, log-rank tests, and Cox regression models with age as the time scale, which implies that age is automatically adjusted for; Cox proportional hazards ratios were calculated as measures for relative risk. Models were left truncated (1980 or at immigration) with delayed entry, and individuals were censored at events, death, permanent emigration, or end of follow-up. Multivariate models were adjusted for age, sex, ethnicity, geographical residency, and educational level. Furthermore, we stratified on age group, sex, ethnicity, geographical residency, educational level, follow-up before and after January 1, 1995, and use of anticoagulants. Analyses stratified on ever versus never use of anticoagulants were restricted to the entire Danish population from 1995 through 2007, with left truncation at 1995. Test of interaction in the Cox multivariate model was performed by introducing a two-factor interaction term.
In case–control study designs we also matched each individual with venous thrombosis to five population control subjects on the basis of age, sex, and year of event. Individuals who died of their venous thromboembolic event were excluded. In an attempt to exclude those with provoked venous thromboembolism, these analyses were done both including and excluding those with venous thromboembolism possibly provoked by surgery, cancer, or pregnancy up until 90 days before a hospitalization for venous thromboembolism. We used the Kaplan-Meier method to calculate median survival times from date of onset of idiopathic interstitial pneumonia and interstitial lung disease, respectively.
We included the entire Danish population in a 27-year follow-up period from 1980 through 2007, comprising 7,418,953 individuals in total. Baseline characteristics are shown in Table 1. We identified 19,557 individuals with idiopathic interstitial pneumonia and 34,493 individuals with interstitial lung disease. The median age at diagnosis for individuals with idiopathic interstitial pneumonia and interstitial lung disease was 69 and 65 years, respectively. The median survival from a diagnosis of idiopathic interstitial pneumonia or interstitial lung disease was 3.0 and 4.1 years, respectively (Figure 1). The corresponding median survival of control subjects matched for age, sex, and year of event was 12.0 and 13.4 years, respectively.
Individuals without Venous Thromboembolism | Individuals with Venous Thromboembolism | |
|---|---|---|
| Age, yr | 27 (1.7–44) | 51 (38–66) |
| Age group, yr | ||
| <20 | 3,358,758 (46%) | 13,130 (8%) |
| 21–60 | 2,957,590 (41%) | 84,182 (53%) |
| 61–70 | 479,055 (7%) | 33,009 (21%) |
| >70 | 464,874 (6%) | 28,355 (18%) |
| Male/female | 3,631,795/3,628,482 | 73,263/85,413 |
| Ethnicity | ||
| Danish | 6,404,646 (89%) | 152,680 (96%) |
| Other | 755,378 (11%) | 5,754 (4%) |
| Geographical residency | ||
| Northern Jutland | 767,266 (11%) | 15,457 (10%) |
| Central Jutland | 1,412,769 (19%) | 25,281 (16%) |
| Funen and southern Jutland | 1,737,395 (24%) | 33,436 (21%) |
| Zealand | 1,031,819 (14%) | 23,784 (15%) |
| Copenhagen | 2,230,717 (31%) | 60,636 (38%) |
| Level of education | ||
| None, including children | 2,266,119 (31%) | 63,475 (40%) |
| Primary school | 1,883,454 (26%) | 46,867 (30%) |
| High school | 355,002 (5%) | 2,285 (1%) |
| Labor work | 1,543,992 (21%) | 29,674 (19%) |
| Short academic education | 211,410 (3%) | 2,699 (2%) |
| Medium academic education | 532,344 (7%) | 7,443 (5%) |
| Long academic education | 331,165 (5%) | 3,329 (2%) |
Figure 1. Kaplan-Meier curves plotting the overall survival after diagnosis according to (left) idiopathic interstitial pneumonia and (right) interstitial lung disease. Shown are the results of a case–control study of individuals with idiopathic interstitial pneumonia and interstitial lung disease in the Danish population from 1980 through 2007, comprising 16,161/31,306 case subjects and 82,317/158,845 control subjects, respectively.
[More] [Minimize]Cumulative incidence by age of idiopathic interstitial pneumonia was higher among those ever diagnosed with venous thromboembolism (log-rank P = 1.5 × 10−155; n = 158,676), pulmonary embolism (P = 3.2 × 10−184; n = 70,586), and deep venous thrombosis only (P = 1.1 × 10−8; n = 88,090) than among control subjects (n = 7,260,277) (Figure 2). Corresponding age-standardized incidence rates were 1.8, 2.8, 1.2, and 0.8 per 10,000 person-years, respectively (Table 2).
Figure 2. Cumulative incidence of (left) idiopathic interstitial pneumonia and (right) interstitial lung disease according to ever-diagnosed venous thromboembolism, pulmonary embolism, and deep venous thrombosis only. The entire Danish population from 1980 through 2007, comprising 7,418,953 individuals, was studied. Values of P are based on log-rank tests.
[More] [Minimize]Idiopathic Interstitial Pneumonia | Interstitial Lung Disease | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Age-standardized Incidence per 10,000 Person-Years | HR (95% CI) | Age-standardized Incidence per 10,000 Person-Years | HR (95% CI) | ||||||||
| n | Age Adjusted | Multivariate Adjusted | Age Adjusted | Multivariate Adjusted | |||||||
| Control subjects | 7,260,277 | 0.70 | 1.0 | 1.0 | 1.50 | 1.0 | 1.0 | ||||
| Venous thromboembolism | 158,676 | 1.60 | 1.8 (1.7–1.9) | 1.8 (1.7–1.9) | 3.25 | 1.7 (1.7–1.8) | 1.7 (1.7–1.8) | ||||
| Pulmonary embolism | 70,586 | 2.49 | 2.5 (2.4–2.7) | 2.4 (2.3–2.6) | 4.33 | 2.2 (2.1–2.4) | 2.2 (2.1–2.3) | ||||
| Deep venous thrombosis | 88,090 | 1.08 | 1.3 (1.2–1.4) | 1.3 (1.2–1.4) | 2.62 | 1.3 (1.3–1.4) | 1.4 (1.3–1.4) | ||||
Multivariate-adjusted hazard ratios for idiopathic interstitial pneumonia were 1.8 (95% CI, 1.7–1.9) in those ever diagnosed with venous thromboembolism, 2.4 (95% CI, 2.3–2.6) in those ever diagnosed with pulmonary embolism, and 1.3 (95% CI, 1.2–1.4) in those ever diagnosed with deep venous thrombosis only, compared with control subjects (Table 2). When stratifying by sex, ethnicity, and geographical residency, we found only minor changes in risk estimates (Table 3). However, when stratifying by age and level of education, we found the highest risk estimates for the youngest age groups and for high level of education. When stratifying by period of follow-up, risk estimates were slightly higher in 1995 through 2007 than in 1980 through 1994.
Idiopathic Interstitial Pneumonia | Interstitial Lung Disease | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Control Subjects | Venous Thromboembolism | Pulmonary Embolism | Deep Venous Thrombosis | Control Subjects | Venous Thromboembolism | Pulmonary Embolism | Deep Venous Thrombosis | |||||
| All | 1.0 | 1.8 (1.7–1.9) | 2.4 (2.3–2.6) | 1.3 (1.2–1.4) | 1.0 | 1.7 (1.7–1.8) | 2.2 (2.1–2.3) | 1.4 (1.3–1.4) | ||||
| Age group, years | ||||||||||||
| <20 | 1.0 | 4.6 (2.6–8.1) | 11 (5.0–22) | 2.5 (1.1–6.2) | 1.0 | 3.7 (2.9–4.7) | 5.8 (4.0–8.5) | 3.1 (2.3–4.1) | ||||
| 21–60 | 1.0 | 2.4 (2.2–2.6) | 3.7 (3.3–4.0) | 1.6 (1.4–1.8) | 1.0 | 2.1 (2.0–2.2) | 3.0 (2.8–3.2) | 1.6 (1.4–1.7) | ||||
| 61–70 | 1.0 | 1.3 (1.2–1.4) | 1.7 (1.5–1.9) | 1.0 (0.8–1.1) | 1.0 | 1.2 (1.1–1.3) | 1.5 (1.4–1.7) | 1.0 (0.9–1.1) | ||||
| >70 | 1.0 | 1.5 (1.3–1.7) | 1.8 (1.5–2.1) | 1.2 (1.0–1.4) | 1.0 | 1.4 (1.3–1.6) | 1.7 (1.5–2.0) | 1.2 (1.0–1.4) | ||||
| Sex | ||||||||||||
| Male | 1.0 | 1.8 (1.6–1.9) | 2.3 (2.1–2.5) | 1.3 (1.2–1.4) | 1.0 | 1.7 (1.6–1.8) | 2.1 (2.0–2.3) | 1.4 (1.3–1.5) | ||||
| Female | 1.0 | 1.8 (1.7–2.0) | 2.6 (2.3–2.9) | 1.3 (1.1–1.4) | 1.0 | 1.8 (1.7–1.9) | 2.4 (2.2–2.6) | 1.4 (1.3–1.5) | ||||
| Ethnicity | ||||||||||||
| Danish | 1.0 | 1.8 (1.7–1.9) | 2.4 (2.3–2.6) | 1.3 (1.2–1.4) | 1.0 | 1.7 (1.7–1.8) | 2.2 (2.1–2.3) | 1.4 (1.3–1.4) | ||||
| Other | 1.0 | 2.1 (1.6–2.9) | 3.4 (2.4–4.9) | 1.2 (0.7–2.0) | 1.0 | 2.0 (1.6–2.5) | 2.8 (2.1–3.7) | 1.4 (1.0–1.9) | ||||
| Geographical residency | ||||||||||||
| Northern Jutland | 1.0 | 1.7 (1.5–2.0) | 2.3 (1.8–2.8) | 1.4 (1.2–1.8) | 1.0 | 1.6 (1.4–1.8) | 2.1 (1.7–2.5) | 1.3 (1.1–1.6) | ||||
| Central Jutland | 1.0 | 1.8 (1.6–2.1) | 2.3 (2.0–2.7) | 1.4 (1.2–1.7) | 1.0 | 1.8 (1.6–2.0) | 2.2 (1.9–2.5) | 1.4 (1.2–1.7) | ||||
| Funen and southern Jutland | 1.0 | 2.0 (1.8–2.2) | 2.6 (2.3–3.0) | 1.5 (1.3–1.7) | 1.0 | 1.9 (1.8–2.1) | 2.5 (2.2–2.7) | 1.5 (1.4–1.7) | ||||
| Zealand | 1.0 | 1.7 (1.5–1.9) | 2.6 (2.1–3.1) | 1.1 (0.9–1.4) | 1.0 | 1.7 (1.5–1.9) | 2.4 (2.1–2.8) | 1.3 (1.1–1.5) | ||||
| Copenhagen | 1.0 | 1.7 (1.5–1.8) | 2.3 (2.1–2.6) | 1.1 (0.9–1.2) | 1.0 | 1.6 (1.5–1.7) | 2.1 (1.9–2.3) | 1.3 (1.1–1.4) | ||||
| Level of education | ||||||||||||
| None, including children | 1.0 | 1.4 (1.3–1.5) | 1.7 (1.6–1.9) | 1.0 (0.9–1.1) | 1.0 | 1.3 (1.2–1.4) | 1.6 (1.5–1.7) | 1.0 (0.9–1.1) | ||||
| Primary school | 1.0 | 2.3 (2.1–2.5) | 3.4 (3.0–3.8) | 1.6 (1.4–1.8) | 1.0 | 2.0 (1.9–2.2) | 2.7 (2.5–3.0) | 1.6 (1.4–1.7) | ||||
| High school | 1.0 | 2.8 (2.5–3.2) | 2.7 (0.9–8.7) | 0.5 (0.1–3.9) | 1.0 | 3.0 (1.9–4.5) | 2.8 (1.4–5.6) | 3.1 (1.8–5.1) | ||||
| Labor work | 1.0 | 2.8 (2.5–3.2) | 4.5 (3.8–5.4) | 1.8 (1.5–2.2) | 1.0 | 2.4 (2.2–2.6) | 3.5 (3.1–3.9) | 1.8 (1.6–2.0) | ||||
| Short academic education | 1.0 | 3.7 (2.3–5.9) | 6.6 (3.7–12) | 2.0 (0.9–4.3) | 1.0 | 2.4 (1.7–3.3) | 4.5 (3.4–5.9) | 1.2 (0.7–2.1) | ||||
| Medium academic education | 1.0 | 3.9 (2.9–5.3) | 7.2 (5.0–10) | 2.1 (1.3–3.4) | 1.0 | 4.5 (3.4–5.9) | 4.5 (3.4–5.9) | 1.7 (1.3–2.4) | ||||
| Long academic education | 1.0 | 3.2 (2.0–5.4) | 5.3 (2.8–9.8) | 1.9 (0.8–4.3) | 1.0 | 3.1 (2.3–4.2) | 4.3 (2.9–6.4) | 2.2 (1.4–3.5) | ||||
| Period of follow–up | ||||||||||||
| 1980–1994 | 1.0 | 2.1 (2.0–2.3) | 2.7 (2.5–3.0) | 1.6 (1.4–1.8) | 1.0 | 2.0 (1.9–2.1) | 2.5 (2.3–2.7) | 1.5 (1.4–1.7) | ||||
| 1995–2007 | 1.0 | 2.4 (2.2–2.7) | 3.6 (3.3–4.1) | 1.3 (1.2–1.4) | 1.0 | 2.2 (2.1–2.4) | 3.0 (2.8–3.2) | 1.4 (1.3–1.4) | ||||
Cumulative incidence by age of the broader diagnosis of interstitial lung disease was higher among those ever diagnosed with venous thromboembolism (log-rank P = 4.4 × 10−170; n = 158,696), pulmonary embolism (p = 342 × 10−218; n = 70,594) and deep venous thrombosis (p = 1.3 × 10−24; n = 88,102), than among control subjects (n = 7,260,542) (Figure 2). Corresponding age-standardized incidence rates were 3.5, 4.8, 2.8 and 1.6 per 10,000 person-years, respectively (Table 2).
Multivariate-adjusted hazard ratios of interstitial lung disease were 1.7 (95% CI 1.7–1.8) in those ever diagnosed with venous thromboembolism, 2.2 (95% CI 2.1–2.3) in those ever diagnosed with pulmonary embolism, and 1.4 (95% CI 1.3–1.4) in those ever diagnosed with deep venous thromboembolism only, compared with control subjects (Table 2). For stratified analyses, the patterns observed for interstitial lung disease were similar to those seen for idiopathic interstitial pneumonia (Table 3).
In these analyses we studied the entire Danish population in a 12-year follow-up period from 1995 through 2007, comprising 6,202,185 individuals in total. Multivariate-adjusted hazard ratios of idiopathic interstitial pneumonia in individuals ever versus never treated with anticoagulants were 1.4 (95% CI, 1.2–1.6) versus 2.8 (95% CI, 2.4–3.1) in those ever diagnosed with venous thromboembolism versus control subjects (venous thromboembolism × anticoagulation use interaction on idiopathic interstitial pneumonia outcome: P = 1.5 × 10−10) (Figure 3). Corresponding hazard ratios in those ever diagnosed with pulmonary embolism were 2.0 (95% CI, 1.7–2.4) versus 4.1 (95% CI, 3.6–4.8) (venous thromboembolism × anticoagulation use interaction on idiopathic interstitial pneumonia outcome: P = 8.5 × 10−9) and in those ever diagnosed with deep venous thrombosis they were 0.9 (95% CI, 0.8–1.2) versus 1.4 (95% CI, 1.1–1.8) (venous thromboembolism × anticoagulation use interaction on idiopathic interstitial pneumonia outcome: P = 0.07). Similar results were found for interstitial lung disease (Figure 3).
Figure 3. Risk of incident (left) idiopathic interstitial pneumonia and (right) interstitial lung disease in those ever diagnosed with venous thromboembolism, stratified by anticoagulant treatment. n = individuals with venous thromboembolism, pulmonary embolism, and deep venous thrombosis only. Values represent multivariate-adjusted hazard ratios. Interaction represents venous thromboembolism × anticoagulation use interaction on idiopathic interstitial pneumonia and interstitial lung disease outcome, respectively.
[More] [Minimize]Idiopathic interstitial pneumonia and interstitial lung disease were diagnosed both before and after a venous thromboembolism event (Figure 4). On average, the time between occurrence of a venous thromboembolism event and a diagnosis of idiopathic interstitial pneumonia and interstitial lung disease was −0.4 (6.2) and −0.2 (6.2) years, respectively [mean (SD)].
Figure 4. Percentage of individuals with various times of diagnosis between venous thromboembolism and (left) idiopathic interstitial pneumonia and (right) interstitial lung disease. Values are in years, counted from time of diagnosis of venous thromboembolism (0 = time of diagnosis of venous thromboembolism).
[More] [Minimize]In an attempt to reduce the influence of reverse causation, that is, that idiopathic interstitial pneumonia could cause venous thromboembolism, rather than vice versa, we here compared individuals with venous thromboembolism, pulmonary embolism, and deep venous thrombosis with population control subjects matched for age, sex, and year of event. Thus, we studied here whether idiopathic interstitial pneumonia and interstitial lung disease were diagnosed after a venous thromboembolism event had occurred.
Multivariate-adjusted hazard ratios for idiopathic interstitial pneumonia were 2.1 (95% CI, 1.9–2.3) in those ever diagnosed with venous thromboembolism, 3.3 (95% CI, 2.9–3.7) in those ever diagnosed with pulmonary embolism, and 1.7 (95% CI, 1.5–1.9) in those ever diagnosed with deep venous thrombosis only, compared with control subjects (Table 4). When excluding those with a discharge diagnosis of surgery, cancer, or pregnancy up until 90 days before hospitalization for venous thromboembolism, we found only minor changes in risk estimates (Table 4).
Idiopathic Interstitial Pneumonia | Interstitial Lung Disease | |||||||
|---|---|---|---|---|---|---|---|---|
| Multivariate-adjusted HR (95% CI) | Multivariate-adjusted HR (95% CI) | |||||||
| n, Case Subjects | n, Control Subjects | All | Excluding Provoked Venous Thromboembolism | All | Excluding Provoked Venous Thromboembolism | |||
| Venous thromboembolism | 134,893 | 699,140 | 2.1 (1.9–2.3) | 2.2 (2.0–2.5) | 2.0 (1.9–2.1) | 2.1 (2.0–2.3) | ||
| Pulmonary embolism | 50,318 | 268,364 | 3.3 (2.9–3.7) | 3.2 (2.7–3.9) | 2.7 (2.5–3.0) | 3.0 (2.6–3.4) | ||
| Deep venous thrombosis | 95,784 | 489,171 | 1.7 (1.5–1.9) | 1.7 (1.4–1.9) | 1.7 (1.6–1.9) | 1.6 (1.5–1.8) | ||
The principal finding of this Danish nationwide cohort study is that ever-diagnosed venous thromboembolism was associated with incident idiopathic interstitial pneumonia, particularly among those never treated with anticoagulants. Also, the risk estimates were highest among those ever diagnosed with pulmonary embolism, among the youngest age groups, and among those with a high level of education.
Importantly, the overall findings were confirmed after exclusion of those individuals likely to have a provoked venous thromboembolism as well as in a matched case–control study where idiopathic interstitial pneumonia was diagnosed after a venous thromboembolism event. Nevertheless, extreme caution needs to be used to convey the interpretation that there is any causal inference that can be derived from this study. There is a clear association between ever having venous thromboembolism and incident idiopathic interstitial pneumonia; however, even the observation that the time of diagnosis of idiopathic interstitial pneumonia in the matched case–control study occurs after a diagnosis of venous thromboembolism does not mean that subclinical lung disease was not present before venous thromboembolism. Certainly, then, idiopathic interstitial pneumonia could also be a risk factor for venous thromboembolism (7). Also, we cannot exclude the possibility that heightened awareness in individuals who are hospitalized could affect the association, that is, a diagnosis of venous thromboembolism could increase the likelihood of also being diagnosed with idiopathic interstitial pneumonia, and vice versa.
Mechanistically, our findings could have a simple and straightforward explanation: many individuals in a lifelong procoagulant state could continuously develop clinically unnoticed small pulmonary emboli leading to progressive lung fibrosis and ultimately idiopathic interstitial pneumonia or interstitial lung disease. In support of this, we previously observed that individuals homozygous for the procoagulant factor V R506Q (factor V Leiden) suffer from severe dyspnea, a cardinal symptom of lung fibrosis, as well as from decreased lung function and increased lung function decline (3). Also, elevated levels of procoagulant mediators may be present in the lungs, where they are involved in pathogenic events leading to idiopathic interstitial pneumonia. Activation of coagulation proteases inside blood vessels is a known event following tissue damage. However, it is also recognized that intraalveolar activation of coagulation proteases with subsequent fibrin deposition plays an important role in lung injury. Over time, the sum of these discrete episodes of injury may lead to progressive lung fibrosis and remodeling of larger areas of the lungs. (8–11). Supporting this notion, anticoagulant therapy appears beneficial in the treatment of lung fibrosis. For example, treatment with thrombin inhibitor attenuates lung collagen deposition in an experimental model of lung fibrosis in rats (12). Similar findings have been reported after intratracheal instillation of activated protein C in mice (13), inhalation of aerosolized heparin in rabbits (14), and intratracheal gene transfer of tissue factor pathway inhibitor in rats (15). A single nonblinded study has shown that anticoagulant therapy is beneficial for the survival in patients with interstitial lung disease (16). Last, our present finding that risk of idiopathic interstitial pneumonia and interstitial lung disease in those ever diagnosed with venous thromboembolism was higher in those never versus ever treated with anticoagulants, indirectly supports the notion that individuals in a lifelong procoagulant state continuously may develop clinically unnoticed small pulmonary emboli and subsequently progressive lung fibrosis and ultimately idiopathic interstitial pneumonia.
In support of our findings, a study by Hubbard and colleagues showed increased risk of deep venous thrombosis in patients with idiopathic interstitial pneumonia (7); that study differed from our nationwide cohort study in that they selected the 920 case subjects and 3,593 control subjects through general practitioners. However, as in our study, in which we used venous thromboembolism as a proxy for a procoagulant state, that study (7) could not deduce whether venous thromboembolism causes idiopathic interstitial pneumonia, or vice versa. Although few studies indicate that anticoagulant therapies may have positive effects on lung fibrosis (12–20), and that a genetic propensity for venous thromboembolism increases the risk for dyspnea and reduces lung function (3), both suggesting that the causation could be from venous thromboembolism to idiopathic interstitial pneumonia, further studies using rigorous intervention design or a mendelian randomization design (21–24) are required to determine the direction of causation between venous thromboembolism and idiopathic interstitial pneumonia.
Our subanalysis and stratified analysis support that the associations we observed are important, and indeed may be even more pronounced than what appeared in the overall analyses. First, that the risk estimates were highest in the youngest age groups strongly support the findings, as comorbidity at older age may partly confound the association. And second, because individuals with a high level of education often are more persistent about getting the right diagnosis and appropriate treatment, the highest risk estimates in those with the highest levels of education adds further validity to our findings.
Limitations of the present study include that these results are mainly for whites of Danish descent, and therefore our results may not necessarily apply to other ethnic groups. However, when we stratified our analysis on ethnic Danes and other ethnicities the results were similar. Another potential limitation of the present study is the lack of ability to adjust for lifestyle risk factors such as smoking, which could be a confounder. However, Hubbard and colleagues found a strong association between deep venous thrombosis and idiopathic interstitial pneumonia even after adjustment for smoking (7). Also, our results show increased risk estimates in individuals never versus ever treated with anticoagulant medication, implying that smoking cannot explain the whole effect of the increased risk estimates. Likewise, we cannot exclude workplace exposure to dust and fumes as potential confounders. Furthermore, we do not know why some individuals with established venous thromboembolism were not treated with anticoagulants, and therefore we cannot exclude that such nontreatment could affect some of the associations seen in the present study. Yet another potential limitation is misclassification of diagnoses. Because we do not know exactly how and why diagnoses might be misclassified, such misclassification could theoretically account for the association (e.g., suppose cryptic dyspnea due to initially undiagnosed idiopathic interstitial pneumonia was attributed erroneously to venous thromboembolism, and only later the correct diagnosis of idiopathic interstitial pneumonia was made). However, in a previous study validation of ICD diagnoses of venous thromboembolism from the national Danish Patients Registry showed that the majority of cases met objective diagnostic criteria and that factor V Leiden heterozygosity and homozygosity were associated, respectively, with a 3-fold and 18-fold risk of venous thromboembolism, using these same ICD diagnoses (2). Also, in the present study we make it highly likely that most people in Denmark given the diagnosis of idiopathic interstitial pneumonia or interstitial lung disease indeed have a correct diagnosis, as we show that the median survival in those given either diagnosis is 3–4 years, similar to that reported previously (1). Nonetheless, idiopathic interstitial pneumonia is difficult to diagnose, the terminology has been poorly defined (25), and only during the past 10–15 years has a consensus of classification of diagnoses been achieved. However, if misclassification of diagnoses exists, this would probably tend to bias risk estimates toward the null hypothesis; that is, a subset of the individuals among those with idiopathic interstitial pneumonia could have even greater risk estimates than those observed, and misclassification therefore is unlikely to explain the strong associations observed in the present study.
In the general population ever-diagnosed venous thromboembolism was associated with idiopathic interstitial pneumonia, particularly among those never given anticoagulant treatment. These findings provide support for the establishment of intervention trials examining the effect of anticoagulant therapy in patients with idiopathic interstitial pneumonia.
Børge G. Nordestgaard initiated the study, which was designed in detail by all four authors. Database handling and statistical analyses were done by Birgitte F. Sode and Sune F. Nielsen, while all four authors contributed to analyses and interpretation of data. Birgitte F. Sode wrote the first draft of the paper, which was scrutinized and finally accepted by the other three authors.
| 1. | Gudmundsson G, Hunninghake G. Interstitial and restrictive pulmonary disorders. In: Rimoin D, Connor JM, Pyeritz RE, Korf BR, eds. Principles and practice of medical genetics, 5th ed. Philadelphia, PA: Churcilll Livingstone Elsevier; 2007. pp. 1431–1444. |
| 2. | Juul K, Tybjaerg-Hansen A, Schnohr P, Nordestgaard BG, Factor V. Leiden and the risk for venous thromboembolism in the adult Danish population. Ann Intern Med 2004;140:330–337. |
| 3. | Juul K, Tybjaerg-Hansen A, Mortensen J, Lange P, Vestbo J, Nordestgaard BG, Factor V. Leiden homozygosity, dyspnea, and reduced pulmonary function. Arch Intern Med 2005;165:2032–2036. |
| 4. | Sode BF, Dahl M, Nielsen SF, Nordestgaard BG. Venous thromboembolism and risk of idiopathic interstitial pneumonia: a nationwide study [abstract, no. A1109]. American Thoracic Society Meeting, May 14–19, 2010. |
| 5. | Pedersen CB, Gøtzsche H, Møller JØ, Mortensen PB. The Danish civil registration system. Dan Med Bull 2006;53:441–449. |
| 6. | Ahmad O, Boschi-Pinto C, Lopez A, Murray C, Lozano R, Inoue M. Age standardization of rates: a new WHO standard. Geneva, Switzerland: World Health Organization; 2001. |
| 7. | Hubbard RB, Smith C, Le JI, Gribbin J, Fogarty AW. The association between idiopathic pulmonary fibrosis and vascular disease: a population-based study. Am J Respir Crit Care Med 2008;178:1257–1261. |
| 8. | Chambers RC, Laurent GJ. Coagulation cascade proteases and tissue fibrosis. Biochem Soc Trans 2002;30:194–200. |
| 9. | Chambers RC. Procoagulant signalling mechanisms in lung inflammation and fibrosis: novel opportunities for pharmacological intervention? Br J Pharmacol 2008;153:S367–S378. |
| 10. | Dabbagh K, Chambers RC, Laurent GJ. From clot to collagen: coagulation peptides in interstitial lung disease. Eur Respir J 1998;11:1002–1005. |
| 11. | Wygrecka M, Jablonska E, Guenther A, Preissner KT, Markart P. Current view on alveolar coagulation and fibrinolysis in acute inflammatory and chronic interstitial lung diseases. Thromb Haemost 2008;99:494–501. |
| 12. | Howell DC, Goldsack NR, Marshall RP, McAnulty RJ, Starke R, Purdy G, Laurent GJ, Chambers RC. Direct thrombin inhibition reduces lung collagen, accumulation, and connective tissue growth factor mRNA levels in bleomycin-induced pulmonary fibrosis. Am J Pathol 2001;159:1383–1395. |
| 13. | Yasui H, Gabazza EC, Tamaki S, Kobayashi T, Hataji O, Yuda H, Shimizu S, Suzuki K, Adachi Y, Taguchi O. Intratracheal administration of activated protein C inhibits bleomycin-induced lung fibrosis in the mouse. Am J Respir Crit Care Med 2001;163:1660–1668. |
| 14. | Gunther A, Lubke N, Ermert M, Schermuly RT, Weissmann N, Breithecker A, Markart P, Ruppert C, Quanz K, Ermert L, et al. Prevention of bleomycin-induced lung fibrosis by aerosolization of heparin or urokinase in rabbits. Am J Respir Crit Care Med 2003;168:1358–1365. |
| 15. | Kijiyama N, Ueno H, Sugimoto I, Sasaguri Y, Yatera K, Kido M, Gabazza EC, Suzuki K, Hashimoto E, Takeya H. Intratracheal gene transfer of tissue factor pathway inhibitor attenuates pulmonary fibrosis. Biochem Biophys Res Commun 2006;339:1113–1119. |
| 16. | Kubo H, Nakayama K, Yanai M, Suzuki T, Yamaya M, Watanabe M, Sasaki H. Anticoagulant therapy for idiopathic pulmonary fibrosis. Chest 2005;128:1475–1482. |
| 17. | Howell DCJ, Goldsack NR, Marshall RP, Gray AJ, McAnulty RJ, Laurent GJ, Chambers RC. Direct thrombin inhibition attenuates bleomycin-induced pulmonary fibrosis. Thorax 1999;54:A16. |
| 18. | Kobayashi H, Gabazza EC, Taguchi O, Wada H, Takeya H, Nishioka J, Yasui H, Kobayashi T, Hataji O, Suzuki K, et al. Protein C anticoagulant system in patients with interstitial lung disease. Am J Respir Crit Care Med 1998;157:1850–1854. |
| 19. | Shimizu S, Gabazza EC, Taguchi O, Yasui H, Taguchi Y, Hayashi T, Ido M, Shimizu T, Nakagaki T, Kobayashi H, et al. Activated protein C inhibits the expression of platelet-derived growth factor in the lung. Am J Respir Crit Care Med 2003;167:1416–1426. |
| 20. | Yasui H, Gabazza EC, Taguchi O, Risteli J, Risteli L, Wada H, Yuda H, Kobayashi T, Kobayashi H, Suzuki K, et al. Decreased protein C activation is associated with abnormal collagen turnover in the intraalveolar space of patients with interstitial lung disease. Clin Appl Thromb Hemost 2000;6:202–205. |
| 21. | Cohen JC, Boerwinkle E, Mosley TH Jr, Hobbs HH. Sequence variations in PCSK9, low LDL, and protection against coronary heart disease. N Engl J Med 2006;354:1264–1272. |
| 22. | Kamstrup PR, Tybjaerg-Hansen A, Steffensen R, Nordestgaard BG. Genetically elevated lipoprotein(a) and increased risk of myocardial infarction. JAMA 2009;301:2331–2339. |
| 23. | Thanassoulis G, O'Donnell CJ. Mendelian randomization: nature's randomized trial in the post-genome era. JAMA 2009;301:2386–2388. |
| 24. | Zacho J, Tybjaerg-Hansen A, Jensen JS, Grande P, Sillesen H, Nordestgaard BG. Genetically elevated C-reactive protein and ischemic vascular disease. N Engl J Med 2008;359:1897–1908. |
| 25. | Ryu JH, Daniels CE, Hartman TE, Yi ES. Diagnosis of interstitial lung diseases. Mayo Clin Proc 2007;82:976–986. |
