Clinically significant extra-cardiac findings in asymptomatic HIV-positive men undergoing cardiac magnetic resonance imaging

Abstract

Increased research-based imaging has led to an increase in clinically significant extra-cardiac findings. HIV patients are at increased risk of having polypathology at a younger age; therefore, it may be hypothesised that they would have more incidental findings on imaging. We reviewed the magnetic resonance imaging results of 169 HIV-positive and 40 HIV-negative, clinically well volunteers undergoing cardiac magnetic resonance imaging scanning to assess the prevalence of subclinical cardiac pathology. This sub-study assessed the prevalence of clinically significant extra-cardiac findings. Associated risk factors were assessed and clinical follow-up and outcome were ascertained. Of the HIV-positive study group, 12/169 (7.1%) vs. 1/40 (2.5%) control patients had a clinically significant extra-cardiac finding which warranted further radiological or clinical intervention (p = 0.28). A total of three out of 169 (1.1%) were highly clinically significant findings. On logistic regression analysis, age was the only significant contributing factor (p = 0.049); no HIV-associated factors were found to be significant. The prevalence of clinically significant extra-cardiac findings of 7.1% in this HIV-positive cohort is comparable to the prevalence found in previous studies carried out on an older, sicker general population. This highlights the need for planning for unexpected outcomes and also the high rate of clinically significant findings in a seemingly well HIV-positive population.

Keywords

HIV AIDS cardiovascular disease cardiac magnetic resonance imaging MRI extra-cardiac findings incidentaloma men Europe

Background

Cardiac magnetic resonance imaging (CMRI) usage is increasing as a non-invasive method of cardiac imaging to detect cardiac abnormalities. It has been approved by the American College of Cardiology Foundation for clinical use in the detection of 17 different cardiac pathologies.¹ The field of vision in CMRI is not only limited to cardiac structures, but the thorax and upper abdomen are also often visualised. With increased utilisation also comes the increasing probability of diagnosing clinically significant extra-cardiac findings (CSECFs); this is inevitable in any radiological screening process. The prevalence of CSECFs reported in previous studies is between 1.1% and 21%.^2–8 Whilst these studies have looked at the prevalence in the general population, no study has looked specifically at an HIV-positive population.

HIV-positive patients are at higher risk of cardiovascular disease and have been found to have comorbidities and polypathology in line with a general population 10 years older.^9–14 The prevalence of incidental imaging findings has been shown to increase with increasing age and therefore one may assume that more ‘incidentalomas’ would be found in cardiac screening of HIV-positive patients.^3,8 One study to date has looked at incidental findings in the HIV-positive population when carrying out cardiac computed tomography (CT) and found a prevalence of clinically significant ECFs of 17%.¹⁵

Methods

This sub-study is part of a larger ongoing study which has the primary objective of identifying subclinical cardiac pathology in clinically well, virally suppressed, HIV-positive men compared to healthy male controls.

This sub-study aimed to determine the prevalence of incidental extra-cardiac findings in this population with characteristic comparisons being made between those identified with CSECFs and those without.

Ethics approval was sought and given by The Adelaide and Meath/St James’s Hospital ethics board. Informed consent was obtained from all patients prior to enrolment.

Between August 2011 and December 2012, 169 asymptomatic, clinically well HIV-positive men, virally suppressed on anti-retroviral therapy (ART) were prospectively recruited for CMRI to assess for structural and functional cardiac disease as part of an ongoing study. Forty HIV-negative, age- and cardiac risk-matched controls were also recruited.

Mean study group age was 46.5 ± 8.7 years vs. 46.0 ± 9.8 in the control group. The HIV-positive group had a mean CD4 cell count of 643 cells/mm³± 243; most were Caucasians, 159 (94.1%), 10 (5.9%) Sub-Saharan Africans. The majority 71.6% (121/169) were men who have sex with men (MSM). Baseline demographics were recorded including CD4 count, HIV viral load, ART history, HIV risk factor, smoking status, BMI, waist:hip ratio, Framingham risk and years since HIV diagnosis (see Table 1).

Table 1.

Study population characteristics, those with incidental findings vs. those without.

	Clinically significant extra-cardiac findings, N = 12 (%)	No clinically significant extra-cardiac findings, N = 157 (%)	p ^b
Age (years)	51.42 ± 10.2	46.08 ± 8.5	0.1
Ethnicity			0.79
Caucasian	12 (100)	147 (93)
Sub-Saharan African		10 (6.4)
HIV acquisition risk			0.43
MSM	8 (66.7)	113 (72)
IVDU	1 (8.3)	3 (1.9)
Heterosexual	3 (25)	32 (20.4)
Other	0 (0)	9 (5.7)
Smoking status			0.97
Current	4 (33.3)	58 (3.9)
Ex-smoker	3 (25)	38 (24.2)
Never smoked	5 (41.7)	61 (38.9)
Smoking pack years^a	16.4 ± 24.9	13.6 ± 17.1	0.55
Current ART			0.64
NNRTI	8 (66.7)	106 (67.5)
PI	4 (33.3)	39 (24.8)
Other		8 (5.1)
Years since HIV diagnosis	8.3 ± 6.1	9.7 ± 6.6	0.51
Framingham risk	11.8 ± 10.8	9.7 ± 7.2	0.58
BMI (kg/m²)	25.9 ± 3.0	25.4 ± 3.5	0.73
Waist:hip ratio	0.92 ± 0.4	0.93 ± 0.75	0.99
Nadir CD4 (cells/mm³)	222 ± 132	254 ± 175	0.55
Current CD4 (cells/mm³)	574 ± 334	649 ± 281	0.17

Among those with a history of cigarette smoking.

p Values calculated with Chi square test for categorical variables and Mann–Whitney U test for continuous variables.

Categorical variables expressed as a percentage and continuous variables as a mean ± SD.

MSM: men who have sex with men; IVDU: intravenous drug use; ART: anti-retroviral therapy; NNRTI: non-nucleoside reverse transcriptase inhibitor; PI: protease inhibitor; BMI: body mass index.

Patients were scanned on a 3 T Philips machine with standard protocols as per Society for Cardiovascular Magnetic Resonance guidelines. They received weight-based intravenous gadolinium. Scans were analysed and reported by a consultant cardiologist experienced in CMRI. Incidental findings were noted on the report and recommended follow-up was communicated to the requesting physician.

Clinically significant findings, for the purposes of this study, were defined as any reported extra-cardiac findings that warranted further follow-up either radiologically or clinically. All patients with clinically significant findings had their medical notes reviewed and the outcome of any incidental findings was recorded.

Statistical analysis was then carried out on SPSS software version 20. Descriptive analysis of the cohort was broken down into those with CSECFs and those without. Comparisons between these two groups were carried out using Chi square test for categorical variables and Mann–Whitney U test for continuous variables. Logistic regression analysis was then carried out to ascertain any significant contributing variables.

Results

A total of 12 out of 169 (7.1%) of the study group vs. 1/40 (2.5%) of the control group, (p = 0.28) had clinically significant non-cardiac findings which required follow-up. Three study patients had major incidental findings: two were newly diagnosed with non-small cell lung cancer (NSCLC), and one with a thymoma, despite all being clinically well and asymptomatic. To date, clinical outcome is known for 7/12 (58%) of the study patients, the other 5/12 are still awaiting follow-up scans, given the time frames of 1–5 years for recommended follow-up scans (see Table 2). The one control patient who required follow-up was the eldest control patient, 69 years. He was found to have a simple liver cyst and was advised to have further evaluation with ultrasound scanning.

Table 2.

MRI finding and follow-up outcome in the 12 patients with CSECFs.

Patient study number	Finding	Follow-up	Diagnosis
8	Mediastinal mass	CT thorax/thymectomy	Thymoma, type A3
14	Multiple liver cysts	Ultrasound liver × 2. MRI liver	Cystadenoma
17	Cystic mass at cardiac border	Chest X-ray	Benign cyst
32	Dilated ascending aorta	Repeat MRI ordered for 1 year	Awaited
50	Lung mass, 5 cm in left upper lobe	CT full body/PET scan full body. Chemotherapy and radiotherapy	Stage 4 non-small cell lung cancer
58	Dilated ascending aorta	Repeat MRI ordered 3–5 years	Awaited
68	Dilated ascending aorta	Repeat MRI ordered 3–5 years	Awaited
76	Liver cyst	Ultrasound liver	Benign cyst
79	Lung nodule	CXR × 2	Benign
102	Dilated ascending aorta	Repeat MRI ordered for 1 year	Awaited
119	Lung mass in right upper lobe	CT thorax, chemotherapy, thoracotomy and radiotherapy.	Stage 3a non-small cell lung cancer
122	Dilated ascending aorta	Repeat MRI ordered for 1 year	Awaited

MRI: magnetic resonance imaging; PET: positron emission tomography; CT: computed tomography.

The most common incidental finding was a dilated ascending aorta in five of 12 individuals, with repeat imaging recommended in 1–5 years. In the remainder thoracic cysts, masses and nodules were identified in five and liver cysts in two patients.

On logistic regression analysis the only significant variant attributable to increased risk of an incidentaloma was age (p = 0.049). No HIV-specific factor was significantly found to contribute.

Outcomes of highly significant clinical findings

Patient 8 was found to have a 5-cm thymic mass. A CT scan of thorax/abdomen and pelvis and a CT-guided biopsy were then carried out. This confirmed the presence of a thymoma. A thymectomy was carried out and a Type A3 thymoma was confirmed with no capsular invasion. He has since made a full recovery.

Patient 50 was found to have a 5-cm mass in the left upper lobe of his lung. Follow-up scans with CT and positron emission tomography revealed brain metastasis, stage 4 disease. Histology from a biopsy confirmed adenocarcinoma. He has subsequently undergone chemotherapy and radiotherapy and continues to be followed by the oncology service.

Patient 119 was found to have a right upper lobe lung lesion. He was diagnosed with stage 3a NSCLC and had neoadjuvant chemotherapy, followed by a thoracotomy and radiotherapy. He subsequently died in December 2013 (see Table 3).

Table 3.

Demographics of the three patients found to have highly significant clinical findings.

Patient	8	50	119
Age (years)	46	53	53
Years since HIV diagnosis	8	2	17
Nadir CD4 cells/mm³	515	273	242
Current CD4 cells/mm³	698	421	1480
HIV VL copies/ml	Not detected	Not detected	Not detected
Smoking history	Non-smoker	Current smoker. 18 pack-year Hx	Ex-smoker 80 pack-year Hx
Additional medical history	Nil significant	Lung decortification for haemopneumothorax in past	Past history of Kaposi’s sarcoma
CMRI finding	Thymoma	Stage 4 lung adenocarcinoma	Stage 3 non-small cell lung cancer

Discussion

This is the first study looking at CSECFs in HIV-positive men undergoing CMRI. We found a prevalence of 7.1% which is slightly lower than the mean prevalence of 9.9% (0.4–21%)^2–8 found in previous, non-HIV specific studies. We did, however, find major pathology in 1.77% of our study cohort, which is greater than the 0.4% and 1.1% of new clinically significant major findings in the largest study of 1534 patients by Chan et al.³ and 714 scans reviewed by Irwin et al.,⁵ respectively.

We did not find any statistically significant difference between our HIV-positive cohort and an age-matched control group. This may be representative of the fact that the numbers were too small or that no significant difference existed. However, we feel that the number and severity of incidentalomas in the study group is clinically significant and would have significant impact on patient care and the clinicians’ work load.

One other study has looked at the prevalence of CSECFs in the HIV population on cardiac CT scans, when looking at coronary artery calcium scores. They found a prevalence of 17% which was similar to the general population rate, albeit in a younger cohort.¹⁵ They also found no significant HIV-related contributing factors; however age, smoking history and a positive coronary artery calcium score increased the chance of finding a CSECF.

The variation in prevalence between CT and MRI pick up of incidentalomas is well recognised. These two modalities are difficult to compare as MRI has lower spatial resolution, meaning that the pickup of smaller lesions is more challenging.² One study compared the two in the same patient cohort and found a prevalence of 5% when CT was carried out and 2% on MRI.⁴

Despite this study being carried out on an HIV-positive cohort we found a lower prevalence in this study compared to studies in the general population. However, it is noteworthy that this study was carried out on well, asymptomatic patients for research purposes, whereas six out of the seven previously published studies have looked for CSECFs in patients having clinically indicated MRI scans.^2,3,5–8 In some instances these patients were intubated, which would increase the chance of finding pathology such as atelectasis, pleural effusions and pneumonias which were often the most common findings. Dewey et al.⁴ found a prevalence of 2% CSECFs in a research group of older patients, mean age 63 years.

Our patient group was a younger cohort than most other studies, mean age 46.5 years vs. 57.0 years.^3,4,6–8 As shown by Atalay et al., and Chan et al., increasing age is a significant risk factor for finding CSECFs. The mean age of those with CSECFs in one group was 66.5 years vs. 51.42 years in our cohort.^2,3 One previous study was comparable by age with a mean of 45 years. However, the indications for CMRI in this group were primarily due to suspected or confirmed cardiac defects; hence, a higher prevalence of non-cardiac findings such as pleural effusions is not unexpected in this group.⁸

The definition of CSECFs varies widely across the studies. This combined with varying fields of vision and MRI pulse sequences lead to difficulties in comparing study prevalence. We feel our definition reflects in a real and meaningful way the impact on patient care.

Most of the published studies to date were carried out retrospectively by radiologists. One study demonstrated that this approach leads to a significant difference in pick up rates of CSECFs, with 11.7% on first reading vs. 21% on second reading.⁷ Our study was based on the actual report at the time, reflecting real life. However, this does highlight the need for radiologists/cardiologists to be vigilant when reading scans to look for incidental findings.

As most studies have been carried out by radiologists and not the primary clinician, the clinical follow-up of the patients and outcomes of the findings was not always available; this may have led to over- or underestimation of the true significance of their findings.^2,6

There are a number of limitations to this study. Firstly, we excluded women from this study, significantly reducing the chance of identifying breast pathology. Secondly, we included a low number of intravenous drug users and co-infected patients. Thirdly, the study cohort was relatively young, asymptomatic, on ART and virally suppressed. All of these factors mean that we may have significantly underestimated the prevalence of CSECFs in the general HIV population. We did not compare these findings to age-matched HIV-negative controls.

An incidental finding in a ‘healthy’ volunteer may have potential health implications for the volunteer; this is often not taken into account when planning a research study. The increasing prevalence of radiological studies and the increased likelihood of finding these incidentalomas in HIV-positive volunteers give us some pause for thought. Guidance on the management of research imaging is inconsistent, limited and does not address the interests of volunteers.¹⁶ When planning for incidental outcomes we need to consider several potential issues. Are incidentalomas discussed in the consent? How and when do we inform the volunteer? Who follows it up, the researcher, the HIV specialist or the general practitioner? Will there be financial, insurance, employment issues? There are also cost implications in the follow-up of false positives for the clinician to take into account. We also must consider who is responsible if a finding outside of the field of research is missed due to the focus solely being on the field in question.

This study will help to add to the knowledge base of expected CSECFs in the HIV population so that researchers and clinicians can plan for unpredictable events. Certainly, there is an onus on radiologists and cardiologists to examine the entire field of vision to ensure that important clinical findings are not missed. We advocate careful considered planning of research projects involving imaging and a discussion around the potential for incidentalomas during the consent process.

Footnotes

Acknowledgements

We would like to acknowledge the Centre for Advanced Medical Imaging in St James’s Hospital, Dublin.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Abbott Virology External Collaborations fund.

References

Douglas

Khandheria

Stainback

. ACCF/ASE/ACEP/AHA/ASNC/SCAI/SCCT/SCMR 2008 appropriateness criteria for stress echocardiography: a report of the American College of Cardiology Foundation Appropriateness Criteria Task Force, American Society of Echocardiography, American College of Emergency Physicians, American Heart Association, American Society of Nuclear Cardiology, Society for Cardiovascular Angiography and Interventions, Society of Cardiovascular Computed Tomography, and Society for Cardiovascular Magnetic Resonance endorsed by the Heart Rhythm Society and the Society of Critical Care Medicine. J Am Coll Cardiol 2008; 51: 1127–1147.

Atalay

Prince

Pearson

. The prevalence and clinical significance of noncardiac findings on cardiac MRI. AJR Am J Roentgenol 2011; 196: W387–W393.

Chan

Smith

Hauser

. Noncardiac pathology on clinical cardiac magnetic resonance imaging. JACC Cardiovasc Imaging 2009; 2: 980–986.

Dewey

Schnapauff

Teige

. Non-cardiac findings on coronary computed tomography and magnetic resonance imaging. Eur Radiol 2007; 17: 2038–2043.

Irwin

Newton

Peebles

. Incidental extra-cardiac findings on clinical CMR. Eur Heart J Cardiovasc Imaging 2013; 14: 158–166.

McKenna

Laxpati

Colletti

. The prevalence of incidental findings at cardiac MRI. Open Cardiovasc Med J 2008; 2: 20–25.

Wyttenbach

Medioni

Santini

. Extracardiac findings detected by cardiac magnetic resonance imaging.[Research Support, Non-U.S Gov't]. Eur Radiol 2012; 22: 1295–1302.

May

Mansfield

Landes

. Prevalence of noncardiac findings in patients undergoing cardiac magnetic resonance imaging. Scientific World Journal 2012; 2012: 474–582.

Guaraldi

Orlando

Zona

. Premature age-related comorbidities among HIV-infected persons compared with the general population. Clin Infect Dis 2011; 53: 1120–1126.

10.

Mathers

Loncar

. Projections of global mortality and burden of disease from 2002 to 2030. PLoS Med 2006; 3: e442–e442.

11.

Crum-Cianflone

Grandits

Echols

. Trends and causes of hospitalizations among HIV-infected persons during the late HAART era: what is the impact of CD4 counts and HAART use? J Acquir Immune Defic Syndr 2010; 54: 248–257.

12.

Klein

Hurley

Quesenberry

Jr . Do protease inhibitors increase the risk for coronary heart disease in patients with HIV-1 infection? J Acquir Immune Defic Syndr 2002; 30: 471–477.

13.

Currier

Taylor

Boyd

. Coronary heart disease in HIV-infected individuals. J Acquir Immune Defic Syndr 2003; 33: 506–512.

14.

Triant

Lee

Hadigan

. Increased acute myocardial infarction rates and cardiovascular risk factors among patients with human immunodeficiency virus disease. J Clin Endocrinol Metab 2007; 92: 2506–2512.

15.

Crum-Cianflone

Stepenosky

Medina

. Clinically significant incidental findings among human immunodeficiency virus-infected men during computed tomography for determination of coronary artery calcium. Am J Cardiol 2011; 107: 633–637.

16.

Booth

Jackson

Wardlaw

. Incidental findings found in “healthy” volunteers during imaging performed for research: current legal and ethical implications. Br J Radiol 2010; 83: 456–465.