Incorporating Global Longitudinal Strain (GLS) into Cardiotoxicity Surveillance

Rationale for Using GLS

GLS detects subclinical myocardial injury earlier than left‑ventricular ejection fraction (LVEF), making it the most sensitive and specific tool for early cardiotoxicity detection. Strong consensus (American Society of Echocardiography & European Association of Cardiovascular Imaging). 1
LVEF measurement is limited by high inter‑observer variability (12–15 %), which hampers reliable detection of the ≥10 % drop used to define cardiotoxicity. Moderate evidence. 2
LVEF declines occur late, often months after irreversible myocardial damage has already taken place. Moderate evidence. 3
GLS shows superior reproducibility (5.5–9.5 % variability) compared with LVEF, enhancing reliability of serial monitoring. Moderate evidence. 2
GLS reflects longitudinal sub‑endocardial fiber function, the myocardial layer that is damaged first in chemotherapy‑related toxicity. Strong consensus. 4

Routine GLS measurement is recommended for all patients at risk for cardiotoxicity; a relative reduction > 15 % from baseline or an absolute GLS > ‑16 % (less negative) should trigger cardioprotective therapy even when LVEF remains normal. Strong consensus (American Society of Echocardiography & European Association of Cardiovascular Imaging). [1][5]
Mild GLS depression is defined as values > ‑16 %; severe reduction as values > ‑10 %. Consensus. 4
Actionable GLS change categories (American Society of Echocardiography & European Association of Cardiovascular Imaging):
- < 8 % reduction from baseline – not clinically meaningful.
- 8–15 % reduction – increase monitoring frequency and consider early cardioprotective therapy.
- > 15 % reduction – very likely abnormal; initiate cardioprotective treatment. Strong consensus. [1][5]3

Baseline assessment should include GLS (2‑D speckle‑tracking from three apical views) and LVEF measured by biplane Simpson’s method or 3‑D echocardiography. Moderate evidence. 2
Anthracycline‑based chemotherapy:
- Measure GLS at cumulative doses of 100–200 mg/m² (early changes detectable).
- Repeat at 200 mg/m² (≥7 % risk of LV systolic dysfunction), 400 mg/m² (≥16 % risk), and 500 mg/m² (≥20 % risk). Strong consensus. 2
HER2‑targeted therapy (e.g., trastuzumab):
- Obtain GLS within 3 months of therapy initiation.
- GLS < ‑19 % predicts subsequent cardiotoxicity and warrants closer surveillance. Strong consensus. 3

Start an ACE inhibitor or beta‑blocker when:
- GLS shows a relative reduction > 15 % from baseline (e.g., from ‑20 % to ‑17 % or worse).
- Absolute GLS becomes less negative than ‑16 %.
- These changes occur even with preserved LVEF. Strong consensus. [1][3]

Use the same ultrasound equipment and software for all serial studies; relative changes within the same patient are reliable despite inter‑vendor absolute‑value differences. Moderate evidence. 2
GLS is not applicable to patients receiving inotropic support or mechanical circulatory support. Consensus. 4

Combining GLS with troponin and BNP markedly increases sensitivity for early cardiotoxicity detection; concurrent abnormalities yield the highest predictive value for later cardiac dysfunction. Strong evidence. 2

The 2019 European Myeloma Network/Italian Society of Arterial Hypertension consensus endorses GLS as the most sensitive technique for early cardiotoxicity detection. Strong consensus. 1
The 2020 European Society for Medical Oncology (ESMO) consensus confirms that GLS facilitates early identification of subclinical dysfunction and may define a therapeutic window before LVEF declines. Strong consensus. 2
While GLS reliably detects early myocardial changes, its predictive value for subsequent clinical cardiotoxicity requires further validation in prospective trials. Limited evidence. [1][3]2

Chemotherapy‑induced myocardial injury often preferentially involves the interventricular septum, a pattern distinguishable on regional longitudinal strain analysis. Consensus. 1

The American Society of Echocardiography (ASE) and the European Association of Cardiovascular Imaging (EACVI) define the normal peak left ventricular global longitudinal strain (GLS) as approximately ‑20 % (range ‑18 % to ‑22 %) with a lower limit of normal (LLN) of ‑16 % (absolute value); values less negative than ‑16 % are considered abnormal. 6, 7, 8, 9
The 2015 ASE/EACVI Chamber Quantification recommendations state that a peak GLS around ‑20 % can be expected in a healthy individual. 6, 8, 9
The LLN of ‑16 % is explicitly reported in the 2017 ASE/EACVI update. 10

Sex: Female subjects tend to exhibit slightly more negative (higher absolute) GLS values than male subjects. 6, 8, 9
Age: GLS magnitude decreases progressively with advancing age, necessitating age‑adjusted interpretation of borderline values. 6, 8

The 2015 ASE/EACVI guidelines caution that inter‑vendor and software differences remain too large to endorse universal normal values for GLS. 6, 8, 9
For serial monitoring, it is recommended to use the same ultrasound equipment, tracing methodology, and software version to minimise variability. 6, 7, 8

Athletes: Normal GLS in athletic hearts ranges from ‑16 % to ‑22 %; values less negative than ‑16 % should raise suspicion for myocardial pathology. 11
In athletes recovering from illness, a GLS greater than ‑17 % (i.e., less negative) warrants evaluation for subclinical ventricular dysfunction. 11
Acute heart failure with preserved ejection fraction (HFpEF):
- GLS < ‑16 % indicates mild depression of myocardial deformation.
- GLS < ‑10 % reflects severe reduction.
- These thresholds require validation in acute settings and are not applicable to patients receiving inotropic support. 12, 13

GLS should be measured from the three standard apical views (four‑chamber, two‑chamber, and three‑plane) and averaged, with the timing reference set at aortic valve closure. 7, 8
Exclude GLS calculations when regional tracking is suboptimal in more than two segments within a single view. 7, 8
Optimise image quality and maximise frame rate to reduce measurement variability. 7, 8
Current evidence supports the use of mid‑wall GLS measurements rather than endocardial strain. 7, 8