Use of a multivariable model to estimate the probability of discharge in hospitalized foals that are 7 days of age or less

Abstract

Combining the clinician's initial predication of the probability of a foal being discharged alive with that of the model appeared to provide a more precise early estimate of the probability of live discharge for hospitalized foals.

Factors associated with outcome and gradual improvement in survival over time in 1065 equine neonates admitted to an intensive care unit

“…Several studies have documented an association between admission lactate concentration and outcome in equine neonates [5,18]. In one study, the clinician's prediction as to whether the foal would survive was as accurate as the best multivariable model [12]. In this study, the odds on nonsurvival were 2.26 times higher in bacteraemic foals and, although not retained in the final model, bacteraemia was significantly (P = 0.006) associated with natural death in the univariable analysis.…”

“…Factors associated with outcome using multivariable analyses in prior studies have varied considerably and include cold extremities, recumbency, suckle reflex, temperature, heart rate, age, gestation <320 days, ≥2 infection sites, IgG, glucose, creatinine, WBC or neutrophils and anion gap [3,4,11,12]. Some of these factors, such as temperature and neutrophils were significantly associated with both nonsurvival and natural death in the current study.…”

“…In addition, bacteraemia, bicarbonate, PCO 2 , presence of infectious orthopaedic disorders, presence of other disorders and sepsis score were significantly associated with nonsurvival in this study. Models to predict outcome in equine neonates have suboptimal predictive values and should not be used as the sole criterion on which to base a decision for euthanasia [11,12]. Lactate was not included in the data analysis in this study because it was only available for foals admitted during the last decade.…”

Prognostic factors in the sick neonatal foal

“…However the studies on the role of blood culture and of multiple foci of infection provided conflicting results (Stewart et al 2002;Sanchez 2005;Peek et al 2006). Other factors significantly related to survival, are: WBC and N count, serum albumin concentration, serum glucose concentration, blood pH, base excess and plasma administration on admission (Gayle et al 1998), standing on admission (Rohrbach et al 2006;Corley et al 2007). In a final multivariate model the foollowing variables have been included: standing, duration of clinical signs prior to admission, respiratory rate, N count and age (Gayle et al 1998).…”

“…Blood culture has been included in a combination of parameters predictive of mortality that retrospectively gave the highest values for Se (100%) and Sp (100%) (Castagnetti, unpublished data). Finally, a mathematical model for the early prediction of survival, including: age, ability to stand, presence of suckle reflex, WBC count, serum creatinine levels and anion gap, provided prospectively PPV 90% and NPV 46% (Rohrbach et al 2006). It is evident that single parameters do not have a high prognostic value, particularly in critically-ill foals referred to intensive care units, and that mathematical models are the best way to increase sensitivity and specificity.…”