Plasma Human Cartilage Glycoprotein-39 Is Associated With the Prognosis of Acute Ischemic Stroke

Abstract

Background To evaluate the prognostic value of plasma YKL-40 (human cartilage glycoprotein-39) for acute ischemic stroke. Methods and Results We measured plasma YKL-40 levels in 3377 participants from CATIS (China Antihypertensive Trial in Acute Ischemic Stroke). Study outcome data on death, major disability (modified Rankin Scale score ≥3), and vascular diseases were collected at 3 months after stroke onset. The primary outcome was defined as a combination of death and major disability. During the 3-month follow-up, 828 participants (24.5%) experienced major disability or died. After multivariate adjustment, the highest YKL-40 quartile was associated with an increased risk of the primary outcome (odds ratio, 1.426 [95% CI, 1.105-1.839]; Ptrend=0.01) compared with the lowest quartile. Each SD increase in log-transformed YKL-40 level was associated with a 15.5% (95% CI, 5.6-26.3%) increased risk of the primary outcome. The multivariable-adjusted spline regression models showed a linear dose-response relationship between YKL-40 and clinical outcomes. Adding YKL-40 to a model containing conventional risk factors significantly improved the reclassification power for the primary outcome (net reclassification improvement: 15.61%, P<0.001; integrated discrimination index: 0.37%, P=0.004) and marginally significantly improved the discriminatory power for the primary outcome (area under the receiver operating characteristic curve improved by 0.003, P=0.099). Conclusions A higher YKL-40 level in the acute phase of ischemic stroke was associated with an increased risk of mortality and major disability at 3 months after stroke, indicating that YKL-40 may play an important role as a prognostic marker of ischemic stroke. Registration URL: https://www.clinicaltrials.gov; Unique identifier: NCT01840072.

Keywords: YKL‐40; acute ischemic stroke; prognosis.

Figures

Figure 1. Association between plasma YKL‐40 levels and adverse clinical outcomes after ischemic stroke.

A, Primary outcome. B, Major disability. C, Death. Odds ratios (red solid line) and 95% CIs (blue dotted line) derived from restricted cubic spline regression, with knots placed at the 5th, 35th, 65th, and 95th percentiles of the distribution of plasma YKL‐40. ORs were adjusted for age, sex, admission NIHSS score, antihypertensive treatment, time from onset to hospitalization, current smoking status, alcohol consumption, body mass index, fasting plasma glucose, dyslipidemia, history of hypertension, history of diabetes, family history of stroke, systolic blood pressure at baseline, and ischemic stroke subtype. NIHSS indicates National Institutes of Health Stroke Scale; OR, odds ratio; and YKL‐40, human cartilage glycoprotein‐39.

Figure 2. Distribution of modified Rankin Scale (mRS) score at 3 months according to plasma YKL‐40 quartiles in patients with ischemic stroke.

YKL‐40 indicates human cartilage glycoprotein‐39.

Figure 3. Subgroup analyses of the association between plasma YKL‐40 and primary outcome (quartile 4 vs quartile 1, data for quartiles 2 and 3 not shown).

The odds ratios and 95% CIs were calculated after adjusting for the same confounding factors in Table 2 unless the variable was used as a subgroup variable. BP indicates blood pressure; OR, odds ratio; and YKL‐40, human cartilage glycoprotein‐39.