fetching data ... 
Background: Hydroxychloroquine (HCQ) cardiotoxicity remains an underrecognized condition. Diagnosis ultimately relies on invasive endomyocardial biopsy (EMB) and non-invasive screening methods are warranted. Strain imaging is a novel tool to detect early subclinical left ventricular (LV) dysfunction and may have a role in screening for HCQ cardiotoxicity (1). Strain measures systolic deformation indices that when decreased can predict cardiovascular outcomes more accurately than LV ejection fraction (2).
Objectives: We assessed whether high HCQ cardiotoxicity risk is associated with a specific strain pattern in a group of patients with SLE and end-stage renal disease (ESRD).
Methods: This was a retrospective study in a tertiary care center in New York on a group of patients with an established diagnosis of SLE, ESRD and cardiomyopathy on the index echocardiogram followed between years 2003 and 2019. The patients were stratified into two groups: high risk HCQ toxicity group was defined as either cumulative HCQ dose ≥1000g and/or an endomyocardial biopsy confirming HCQ toxicity. Low/moderate risk group was defined as a cumulative dose of HCQ <1000g. Clinical, demographic, electrocardiographic and echocardiographic strain parameters were compared between the groups.
Results: A total of 16 patients were included. Two patients had EMB consistent with HCQ induced toxicity and 3 patients had cumulative HCQ doses ≥1000g. There were no significant differences in the baseline demographic characteristics between the two groups. Compared to patients with low/moderate risk, patients in the high risk group had a lower heart rate at the time of the echocardiogram (69 vs 87 beats per minute, p=0.08) and a higher frequency of LV hypertrophy (40% vs 9.1%, p=0.2). Strain analysis showed that both groups had compromised LV global longitudinal strain (GLS) and global cross-sectional strain (GCS). However, compared to the low/moderate risk group, the high risk group had a weaker LV GLS (-12.3% vs -14.9%, p=0.27).
Characteristics overall and stratified by HCQ risk group
| Characteristic | Overall (n=16 ) | Low/Moderate HCQ Risk(n=11 ) | High HCQ Risk and/or Positive EMB (n=5 ) | P value | |
|---|---|---|---|---|---|
| Demographics | Age, years | 47.5 (36.5,60.7) | 50.0 (33.9,60.5) | 42.5 (42.7,61.0) | 0.95 |
| Female, n(%) | 14 (87.5) | 9 (90) | 5 (83.3) | 0.99 | |
| Clinical Features | SLE duration, years | 7.4 (4.3,17.5) | 5.5 (3.5,13.2) | 15.6 (11.6,19.3) | 0.15 |
| HCQ cumulative dose, g | 422.8 (224.2,422.8) | 285.4 (110.8,523.6) | 1140 (1006,1625.4) | 0.005 | |
| HCQ therapy duration, years | 3.4 (2.5, 8.9) | 3.2 (1.5,5.1) | 7.8 (6.8,11.6) | 0.06 | |
| HCQ daily dose, mg/d | 226 (200,394.9) | 200 (179.4,253.7) | 400 (389.8,400) | 0.007 | |
| Hypertension, n (%) | 14 (87.5) | 10 (90.9) | 4 (80.0) | 0.99 | |
| Diabetes, n (%) | 3 (18.8) | 2 (18.8) | 1 (20.0) | 0.99 | |
| CAD, n (%) | 3 (18.8) | 2 (18.8) | 1 (20) | 0.99 | |
| Echocardiogram | EF, % | 55 (42.5,60) | 55 (40,60) | 55 (55,70) | 0.45 |
| LA size, cm | 3.8 (3.4,4.3) | 3.8 (3.4,4.2) | 4.3 (3.4,4.9) | 0.30 | |
| LVEDD, cm | 4.9 (4.4, 5.5) | 4.8 (4.2,5.5) | 5.0 (4.9,5.4) | 0.43 | |
| E/E’ | 12.3 (8.8,16.3) | 12 (8.8,14.9) | 16.9 (4.9,21.8) | 0.43 | |
| Moderate-severe LV hypertrophy, n(%) | 3 (18.7) | 1 (9.1) | 2 (40.0) | 0.20 | |
| Strain echocardiography | GLS, % | -13.9 (-16.7,-12.3) | -14.9 (-16.7,-12.9) | -12.3 (-15.4,-12.2) | 0.27 |
| Base/Apex Ratio | 0.8 (0.7,0.9) | 0.76 (0.68,0.86) | 0.76 (0.66,0.83) | 0.95 | |
| GCS, % | -20.2 (-21.7,-17) | -19.7 (-20.5,18.0) | -21.7 (-23.9,-20.9) | 0.16 | |
| RV GLS, % | -20.19 (-22.1,-17.5) | -20.2 (-22.3,-17.1) | -19.8 (-22.1,-17.5) | 0.99 |
Conclusion: We report an association of higher HCQ cardiotoxicity risk and impaired strain in a set of SLE ESRD patients. Standard echo measures did not differentiate between high and low/moderate risk patients. Although the findings did not reach statistical significance, given the small sample size, results are still suggestive of a possible utility of strain echocardiography for detection of early HCQ toxicity.
REFERENCES:
[1]Buss SJ, et al. J Rheumatol. 2010;37(1):79-86
[2]Kalam K, et al. Heart. 2014;100(21):1673-80