The association of capillary refill time and return of spontaneous circulation during out-of-hospital cardiac arrest: an observational study

Abstract

Microcirculatory alterations are predictive of poor outcomes in patients with shock and after cardiac arrest in animal models. However, microcirculatory alterations during human cardiac arrest have not yet been studied. We prospectively included adult patients receiving resuscitation after witnessed out-of-hospital cardiac arrest. Exclusion criteria were hypovolemia, hypo- or hyperthermia (< 34.0 °C, > 37.5 °C), peripheral arterial disease, Raynaud’s disease, and logistical issues (e.g., shortage of space). Capillary refill time was measured on the finger (CRT-F) and the earlobe (CRT-E) every other minute until return of spontaneous circulation (any ROSC) or termination of resuscitation. The primary endpoint was any ROSC, secondary endpoints were 30-day-mortality and good neurological outcome (defined as cerebral performance category 1–2). Based on the data structure, CRT-F and CRT-E values were grouped post-hoc into quartiles and tertiles. A cluster-robust standard error logistic regression was performed for the primary outcome. Trend analyses were made for each individual. After screening of 141 patients, 50 were included in the analysis (median age 75 years, 28% female, any ROSC 32%). The median CRT-F was > 10 [7–> 10] seconds; the median CRT-E was 3 [3–4] seconds. The any ROSC rate for patients in CRT-F quartile 1 (3–5 s) was 71.4%, 31.7% in quartile 2 (6–8 s), 23.1% in quartile 3 (9–10 s), and 10% in quartile 4 (> 10 s). The odds ratio of 0.39 (95% CI 0.20–0.73, p = 0.004) indicated, that with an increase of CRT-F by a quartile, the chance of achieving any ROSC decreased by 61%. Patients with a decreasing CRT-F achieved any ROSC in 70%, whereas patients with constant or increasing CRT-F had any ROSC in only 21% (p = 0.008). In contrast, CRT-E showed no association with any ROSC (T1 [1–2 s.]: 16.7%, T2 [3 s.]: 27.5%, T3 [4—> 10 s.]: 22.4%, OR by tertiles: 1.18, 95% CI 0.58–2.44, p = 0.646). During out-of-hospital cardiac arrest, shorter CRT-F, but not CRT-E, is associated with a higher chance of any ROSC. Trial registration: This trial was pre-registered on clinicaltrials.gov with the number: NCT04791995 on March 2nd, 2021.