Injection drug users (IDUs) account for a considerable number of the hospitalizations for infective endocarditis (IE), but the prevalence of diagnosed and unrecognized IE in IDUs is unknown. The aim of the present study was to assess the prevalence of valvular abnormalities suggestive of IE in IDUs attending a supervised injection facility. We performed transthoracic echocardiographic examinations on-site in the injection facilities. A total of 206 IDUs (mean age 43 ± 9 years, 23% women) with a median injection drug abuse of 18 years (interquartile range 10 to 26) were included. Fourteen IDUs (14 of 206, 7%, 95% confidence interval [CI] 4% to 11%) had a previous history of IE. IDUs with a history of IE were significantly older than IDUs without a history of IE (48 ± 8 vs 42 ± 9 years, respectively, p = 0.03) and had a longer duration of injection drug use (27 [18 to 36] vs 17 years [10 to 25], p = 0.008). In the subgroup of IDUs with a history of IE, 4 subjects (4 of 14, 29%, 95% CI 11% to 55%) had persistent or relapse vegetations. Of the remaining 10 IDUs with a history of IE, 5 (5 of 10, 50%, 95% CI 24% to 76%) had moderate-to-severe regurgitation. In the subgroup of IDUs without a history of IE, vegetations were seen in 9 subjects (9 of 192, 5%, 95% CI 2% to 9%). This group of IDUs with possibly unrecognized IE was older than IDUs without vegetations (48 ± 12 vs 42 ± 9, respectively, p = 0.04). Among the IDUs without a history of IE who did not have vegetations, 30 IDUs (30 of 183, 16%, 95% CI 11% to 22%) had moderate-to-severe regurgitation with or without concomitant thickening of leaflets. Thus, in IDUs without a history of IE, some extent of valvular abnormalities was seen in 20% (39 of 192, 95% CI 15% to 27%) of subjects. None of the IDUs with valvular vegetations had current symptoms consistent with active IE. In conclusion, valvular abnormalities assessed by echocardiography were prevalent in asymptomatic IDUs without a medical history of IE, and vegetations were seen in 5% of subjects.
The incidence of infective endocarditis (IE) in the general population is 0.03 to 0.1/1,000 years. Injection drug use is a major risk factor for IE. The Danish Health and Medicines Authority estimates the number of injection drug users (IDUs) in Denmark to be 13,000. In surveys of IE, drug abuse has been related to 6% to 64% of hospitalizations for IE. The prevalence of IE among active, nonhospitalized IDUs is, however, unknown. Previous estimates of the incidence of IE among IDUs are wide ranging from 0.7 to 13/1,000 years. The estimates are based on studies in highly selected populations of IDUs or on assumptions of the total number of IDUs in the community.
Diagnosing IE is often challenging because of the variable clinical presentation of the disease not least in right-sided IE. Consequently, it is important to be aware of both the prevalence of underlying valvular heart disease predisposing IE and the frequency of IE among IDUs when assessing patients with suspected IE and injection drug abuse. Echocardiography and microbiological findings comprise the cornerstones in the diagnosis of IE according to the modified Duke criteria.
Supervised injecting facilities are legal, public facilities for hygienic consumption of drugs under supervision of health care professionals. Denmark’s first supervised injection facility opened in 2011 and in 2013 The Danish Health and Medicines Authority reported of approximately 240 visits daily.
The aim of the present study was to assess the prevalence of valvular abnormalities suggestive of IE in IDUs attending supervised injection facilities.
Methods
The present study is a cross-sectional, observational echocardiographic study of IDUs in Copenhagen, Denmark. IDUs were included in the study during a 2-week period in October 2012 when attending 1 of the 5 different supervised injection facilities. Inclusion criteria were current injection drug abuse and intake at 1 of the supervised injection facilities during the study period. All IDUs attending the injection facilities during the study period were orally informed about the study and asked to participate. The echocardiographic examination took place on-site in the injection facility in immediate association with the injection of drugs at the facility. Data on age, sex, duration of injection drug abuse, and previous heart disease were collected by interview. Clinical symptoms were recorded by interview in IDUs with echocardiographic signs of IE. According to the study protocol, further examinations were recommended if IDUs with echocardiographic signs of IE had symptoms consistent with ongoing IE. The study was performed in accordance with the Declaration of Helsinki and with local ethical regulations. No formal approval from the Danish National Committee on Health Research Ethics was required because of the descriptive and anonymous design of the study. Data were collected anonymously. All participants gave oral informed consent.
A focused transthoracic echocardiographic examination was performed by experienced operators on-site in the injection facility using Philips CX50 ultrasound system (Philips, Best, The Netherlands). Two-dimensional images from standard apical, parasternal, and subxiphoid views as well as color Doppler recordings of valves were obtained. Modified, off-axis projections to optimize visualization of the tricuspid valve were obtained when needed.
Echocardiographic examinations were analyzed by the operator on-site and stored digitally for a second, systematic off-line interpretation by an experienced cardiologist using Philips Xcelera analysis software version 3.1 (Philips Healthcare, Best, The Netherlands). The cardiologist was blinded to medical history and the result of the primary interpretation. Valvular abnormalities were reported for each valve individually. Vegetation was defined as an oscillating or nonoscillating mass on valves or other endocardial structures. Regurgitation was reported when moderate or severe. In case of interobserver disagreement between the operator and the cardiologist, a second cardiologist reviewed the images and reached a final decision.
The modified Duke criteria are widely used and accepted diagnostic criteria for IE and are based on the presence of multiple findings associated with IE including microbiological and clinical findings as well as echocardiographic findings suggestive of endocardial involvement. Although the criteria are not designed to assess signs of endocardial involvement in a population without clinical suspicion of IE, we applied the echocardiographic components of the criteria in our population to enable comparison with other studies. The remaining criteria based on microbiological or clinical findings could not be applied. According to the echocardiographic signs in the modified Duke criteria, we defined possible IE as the presence of a vegetation, abscess, or new partial dehiscence of a prosthetic valve (major criteria) in addition to the known injection drug use (minor criterion) and definite IE as the presence of 2 of the above-mentioned major criteria. Moderate-to-severe regurgitation was reported but not included in the definition of possible IE as regurgitation could not be defined as new ( Table 1 ).
| Major criteria |
|---|
Blood cultures positive for IE:
|
| Evidence of endocardial involvement Echocardiogram positive for IE defined as follows:
|
| Minor criteria |
|---|
|
Statistical analysis was performed using SAS version 9.2 (Cary, North Carolina). Normally distributed data are presented as mean ± SD and non-normally distributed data as median (interquartile range). Student t and chi-square or Fischer’s exact test when appropriate were used for the comparison of continuous and categorical variables, respectively. Proportions are presented with 95% confidence intervals (CIs). A two-sided p value of <0.05 was considered statistically significant.
Results
Of a total of 251 IDUs attending the injection facilities during the study period, 206 IDUs consented to participate and were included in the study. A history of heart disease was reported by 23 subjects (11%). Fourteen subjects (7%) had a previous history of IE including 2 subjects (1%) who had previously undergone aortic valve replacement due to IE. IDUs with a history of IE were older than IDUs without a history of IE (48 ± 8 vs 42 ± 9, respectively, p = 0.03) and had a longer duration of injection drug use (27 [18 to 36] vs 17 [10 to 25], p = 0.008). There was no significant difference in the prevalence of previous IE between men and women (5% vs 13%, p = 0.09). Demographic characteristics and self-reported heart disease are listed in Table 2 .
| Variable | (n = 206) |
|---|---|
| Age (years) | 43 ± 9 |
| Duration of injection drug abuse (years) | 18 (10–26) |
| Women | 47 (23%) |
| Self-reported previous heart disease | |
| Infective endocarditis | 14 (7%) |
| Pulmonary embolus | 1 (1%) |
| Atrial fibrillation | 1 (1%) |
| Heart failure | 1 (1%) |
| Angina pectoris | 1 (1%) |
| Long QT | 1 (1%) |
| Previous cardiac arrest | 1 (1%) |
| Unspecified heart disease | 4 (2%) |
In the subgroup of IDUs with a history of IE, 4 subjects (4 of 14, 29%, 95% CI 11% to 55%) had persistent or relapse vegetations, 2 on native tricuspid valves, 1 on a native mitral valve, and 1 on a prosthetic aortic valve. The latter, a 62-year-old man with a 42-year history of injection drug use and a biological prosthetic aortic valve due to a previous episode of IE patient, had a 3-mm thick oscillating vegetation on the prosthetic valve. There was no dehiscence or paravalvular regurgitation of the valve. Of the remaining 10 IDUs with a history of IE, 5 (5 of 10, 50%, 95% CI 24% to 76%) had moderate-to-severe regurgitation. Thus, in the 14 IDUs with a history of IE, valvular abnormalities were seen in 64% (9 to 14, 95% CI 39% to 84%) of subjects. Echocardiographic findings are listed in Table 3 .
| Variable | IDUs Meeting Criteria for Possible IE (n = 13) | IDUs Not Meeting Criteria for Possible IE (n = 193) | Total Population (n = 206) |
|---|---|---|---|
| Aortic valve | |||
| Vegetation | 1 (8%) | 0 | 1 (1%) |
| Moderate-to-severe regurgitation | 0 | 3 (2%) | 3 (1%) |
| Thickened leaflets | 0 | 2 (1%) | 2 (1%) |
| Prosthetic valve | 1 (8%) | 1 (1%) | 2 (1%) |
| Mitral valve | |||
| Vegetation | 2 (15%) | 0 | 2 (1%) |
| Moderate-to-severe regurgitation | 1 (8%) | 14 (7%) | 15 (7%) |
| Thickened leaflets | 2 (15%) | 6 (3%) | 8 (4%) |
| Pulmonary valve | |||
| Vegetation | 0 | 0 | 0 |
| Moderate-to-severe regurgitation | 1 (8%) | 7 (4%) | 8 (4%) |
| Thickened leaflets | 0 | 0 | 0 |
| Tricuspid valve | |||
| Vegetation | 10 (77%) | 0 | 10 (5%) |
| Moderate-to-severe regurgitation | 10 (77%) | 25 (13%) | 35 (17%) |
| Thickened leaflets | 4 (31%) | 7 (4%) | 11 (5%) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
