The diagnostic accuracy of clinical examination for estimating cardiac index in critically ill patients: the Simple Intensive Care Studies‑I
Clinical examination has classically a primary role in the diagnosis of low cardiac output state, as it is rapid and cost effective. However, little is known about the diagnostic performance of clinical signs to detect low cardiac output, as previous studies have not systematically investigated this aspect.
Hiemstra et al conducted the SICS study to establish which clinical signs or combination of signs was associated with cardiac index, and their performance in discrimination of low and high cardiac output states.
The authors conducted a single-centre prospective, observational study, including a large unselected cohort of unplanned ICU admission. Patients were evaluated according to 19 clinical variables, and their findings compared with Critical Care Cardiac Ultrasound (CCUS).
The final cohort included 783 patients, with 280 patients (36%) having low cardiac index (<2.2 L min-1 m-2), as detected by CCUS. In the overall population, the three “windows” on organ perfusion were frequently abnormal: skin perfusion was altered in 76% patients, renal perfusion in 68% patients, and mentation in 26% patients, while all three signs were normal in only 5% of patients.
In univariate analysis, signs of reduced perfusion were significantly associated with cardiac index. In multivariable linear regression, 7 out of 19 clinical signs (respiratory rate, heart rate and rhythm, systolic and diastolic blood pressure, central to peripheral temperature [ΔTc–p], and capillary refilling time [CRT]) were independently associated with cardiac index.
When considering diagnostic accuracy, no single finding had high sensitivity, NPV or PPV, while specificity was over 90% for 5 signs only: low systolic blood pressure (<90 mmHg), low diastolic pressure (<45 mmHg), CRT>4.5s at sternum, knee mottling, altered consciousness, and atrial fibrillation. However, only 40% of patients with low cardiac output has an alteration in one of these 5 signs.
Multivariable logistic regression, showed that respiratory rate, heart rate, atrial fibrillation, and systolic/diastolic blood pressure, CRT and ΔTc–p were independently associated with low cardiac output, and able to correctly classify 71% of patients.
The study concludes that physicians should not rely solely on clinical examination for decision making in critically ill patients, and CCUS (a rapid and cost-effective tool) should be complemental to guide the therapeutic strategy.
STUDY STRENGTHS & LIMITATIONS
This study is the first to assess diagnostic accuracy of a large set of clinical signs in a population of ICU patients, and to compare it with CCUS. It gives a strong insight on the reliability of findings used in clinicians’ routine practice.
Study limitations are related with observational design (i.e.: control for confounders)
TAKE HOME MESSAGE
Clinical signs have low diagnostic accuracy for low cardiac index.
Only 5 signs (reported above), have good specificity, and may be used to concluded that a low cardiac output is likely when they are present.
CCUS should supplement clinical examination to guide therapeutic strategies in critically ill patients. Is it time to add a fifth pillar to bedside Physical examination: Inspection, Palpation Percussion Auscultation and, Insonation…
This article review was prepared and submitted by Paul Abraham and Massimiliano Greco, on behalf of the EJRC.
1) Hiemstra B, Koster G, Wiersema R, et al. The diagnostic accuracy of clinical examination for estimating cardiac index in critically ill patients: the Simple Intensive Care Studies-I. Intensive Care Med. 2019 Feb;45(2):190-200.