Functional abilities, respiratory and cardiac function in a large cohort of adults with Duchenne muscular dystrophy treated with glucocorticoids

Abstract

Background and purpose: The transition to adult services, and subsequent glucocorticoid management, is critical in adults with Duchenne muscular dystrophy. This study aims (1) to describe treatment, functional abilities, respiratory and cardiac status during transition to adulthood and adult stages; and (2) to explore the association between glucocorticoid treatment after loss of ambulation (LOA) and late-stage clinical outcomes.

Methods: This was a retrospective single-centre study on individuals with Duchenne muscular dystrophy (≥16 years old) between 1986 and 2022. Logistic regression, Cox proportional hazards models and survival analyses were conducted utilizing data from clinical records.

Results: In all, 112 individuals were included. Mean age was 23.4 ± 5.2 years and mean follow-up was 18.5 ± 5.5 years. At last assessment, 47.2% were on glucocorticoids; the mean dose of prednisone was 0.38 ± 0.13 mg/kg/day and of deflazacort 0.43 ± 0.16 mg/kg/day. At age 16 years, motor function limitations included using a manual wheelchair (89.7%), standing (87.9%), transferring from a wheelchair (86.2%) and turning in bed (53.4%); 77.5% had a peak cough flow <270 L/min, 53.3% a forced vital capacity percentage of predicted <50% and 40.3% a left ventricular ejection fraction <50%. Glucocorticoids after LOA reduced the risk and delayed the time to difficulties balancing in the wheelchair, loss of hand to mouth function, forced vital capacity percentage of predicted <30% and forced vital capacity <1 L and were associated with lower frequency of left ventricular ejection fraction <50%, without differences between prednisone and deflazacort. Glucocorticoid dose did not differ by functional, respiratory or cardiac status.

Conclusion: Glucocorticoids after LOA preserve late-stage functional abilities, respiratory and cardiac function. It is suggested using functional abilities, respiratory and cardiac status at transition stages for adult services planning.

Keywords: EK scale in adults with DMD; cardiac function in adults with DMD; glucocorticoid dose in adults with DMD; respiratory function in adults with DMD; transition to adulthood in DMD.

FIGURE 1

(a) Glucocorticoid type and regimen prescribed at four time points of follow‐up: initial prescription, assessment prior to LOA, LA after LOA on GC and LA. (b) Variations in GC prescription after LOA. To label an increment or decrement on GC dose after LOA, an increment or reduction of ≥25% of the GC dose (mg/kg/day) was considered, as suggested by the 2018 Standards of Care in DMD [4]. Abbreviations: GC, glucocorticoid; LA, last assessment; LOA, loss of ambulation. Last assessment after LOA on GC signifies that the glucocorticoid status of all individuals was evaluated at the point when those who were on glucocorticoids (and subsequently stopped them at the last assessment) received this treatment for the last time.

FIGURE 2

The bar graphs depict scores on the 17 domains of the Egen Klassifikation scale version 2 at the age of 16 years, 18 years and at last assessment. All individuals were non‐ambulant. Darker colours indicate higher scores on the EK scale domains, signifying poorer performance. Domains in blue represent motor function abilities associated predominantly with lower limb strength; in grey motor function abilities associated predominantly with axial strength; in orange motor function abilities associated predominantly with upper limb strength; in green abilities associated predominantly with respiratory muscle strength; and in red abilities associated predominantly with oropharyngeal muscle strength. Within each colour group, domains are arranged in order of impairment. The first EK scale publication was in 2001 [25], so EK scale data at age 16 years old were available for individuals born since 1985.

FIGURE 3

Cox proportional hazards model and adjusted survival analysis comparing time to late‐stage functional ability milestones by glucocorticoid status after loss of ambulation. (a) Cox proportional hazards model and adjusted survival analysis to time to balance in a wheelchair with limitations by glucocorticoid status after loss of ambulation. Balance in a wheelchair with limitations was defined as scoring 2 or 3 on the ‘Ability to balance in the wheelchair’ domain of the Egen Klassifikation version 2 at last assessment. Restricted mean survival time was defined as the area under the curve of the survival function up to a time (time period after LOA 0–13.51 years). (b) Cox proportional hazards model and adjusted survival analysis to time to loss of hand to mouth function by glucocorticoid status after loss of ambulation. Loss of hand to mouth function was defined as scoring 2 or 3 on the ‘Ability to move the arms’ domain of the Egen Klassifikation version 2 at last assessment. Restricted mean survival time was defined as the area under the curve of the survival function up to a time (time period after LOA 0–12.35 years). Note: Dotted lines indicate the median survival time by glucocorticoid status after LOA, except for DFZ as its probability was never 50% or less. The Cox proportional hazards model had two model specifications, which differ only by the baseline glucocorticoid reference group (naïve or prednisone). 95% CI, 95% confidence interval lower bound–upper bound. Abbreviations: DFZ, deflazacort; GC, glucocorticoid; HR, hazard ratio; LOA, loss of ambulation; PDN, prednisone/prednisolone; RMST, restricted mean survival time; SE, standard error.

FIGURE 4

The bar graphs describe the respiratory profile and use of ventilatory support at three time points of follow‐up. Numbers in parentheses represent the number of individuals in each group. (a) Distribution of individuals by FVCpp. Individuals with an FVCpp <50% are coloured in red. Individuals with an FVCpp ≥50% are coloured in blue. (b) Distribution of individuals by FVC <1 L (dark colour) or ≥1 L (light colour). (c) Distribution of individuals by PCF ≤270 L/min (dark colour) or >270 L/min (light colour). (d) Distribution of individuals by NIV user (dark colour) or non‐user (light colour). Note: The GC‐naïve group includes the 10 individuals who discontinued glucocorticoids before or at loss of ambulation. The three individuals on NIV 24 h and the four individuals on tracheostomy were labelled PCF ≤270 L/min and FVC ≤1 L at LA. Distribution of FVCpp at LA by GC type at LA: prednisone <30%, 13; 30%–50%, 10; 50%–80%, four; >80%, no individuals; deflazacort <30%, 24; 30%–50%, 10; 50%–80%, nine; >80%, three individuals. Abbreviations: FVC, forced vital capacity; FVCpp, forced vital capacity percentage of predicted; LA, last assessment; NIV, non‐invasive ventilation; PCF, peak cough flow.

FIGURE 5

Cox proportional hazards model and adjusted survival analysis comparing time to late‐stage respiratory and cardiac milestones by glucocorticoid status after loss of ambulation. (a) Cox proportional hazards model and adjusted survival analysis comparing time to a FVCpp below 30% by glucocorticoid status after loss of ambulation. Restricted mean survival time was defined as the area under the curve of the survival function up to a time (time period after LOA 0–16.10 years). (b) Cox proportional hazards model and adjusted survival analysis comparing time to a LVEF below 40% by glucocorticoid status after loss of ambulation. Restricted mean survival time was defined as the area under the curve of the survival function up to a time (time period after LOA 0–27.10 years). Note: Dotted lines indicate the median survival time by glucocorticoid status after LOA. The Cox proportional hazards model had two model specifications, which differ only by the baseline glucocorticoid reference group (naïve or prednisone). 95% CI, 95% confidence interval lower bound–upper bound. Abbreviations: DFZ, deflazacort; FVCpp, forced vital capacity percentage of predicted; GC, glucocorticoid; HR, hazard ratio; LOA, loss of ambulation; LVEF, left ventricular ejection fraction; PDN, prednisone/prednisolone; RMST, restricted mean survival time; SE, standard error.

FIGURE 6

The bar graphs describe the cardiac function at three time points of follow‐up. Numbers in parenthesis represent the number of individuals in each group. Heart failure (HF) was classified as per the European Society of Cardiology 2021: HF with reduced LVEF (LVEF <40%); HF with mildly reduced EF (LVEF 41%–49%) and HF with preserved EF (LVEF >50%). The GC‐naïve group includes the 10 individuals who discontinued glucocorticoids before or at LOA. Abbreviations: GC, glucocorticoids; LOA, loss of ambulation; LVEF, left ventricular ejection fraction.