Comment Authors: Daniela Pimenta Silva, Diana Aguiar de Sousa, Department of Neurology, Hospital de Santa Maria, University of Lisbon, Portugal
Original Article: Skafida A., Mitrakou A., Georgiopoulos G., et al. In-hospital dynamics of glucose, blood pressure and temperature predict outcome in patients with acute ischaemic stroke. European Stroke Journal. doi: 10.1177/2396987318765824

Stroke unit treatment is an evidence-based proven effective treatment. Surprisingly, randomised controlled trials aimed at reducing or normalising parameters such as glucose, blood pressure (BP) and temperature failed to demonstrate benefit from these interventions. But why wouldn’t the normalisation of these parameters improve outcomes?

Skafida and colleagues hypothesized that the variability of these parameters during the acute and sub-acute phase are significant factors in outcome.

To prove this argument, the authors conducted a prospective observational study, which included 1271 patients with acute ischaemic stroke admitted within 24h after symptom onset to the acute stroke unit of Alexandra Hospital, between 2001 and 2010. Serum glucose, systolic and diastolic blood pressure (SBP, DBP) and temperature were systematically measured during the first 7 days of hospitalisation. Variability across the hospitalisation period was addressed.

The primary outcome was all-cause death up to three months after acute iscaemic stroke and the secondary outcome was poor functional outcome also at three months. Final multivariable models were adjusted for all available major confounders of biological plausibility, including traditional risk factors, stroke severity, renal function and in-hospital treatment.

The authors found that subject-specific baseline glucose (HR=1.005; p-value=0.017) and temperature (HR=2.758; p-value<0.001) levels, variability of SBP (HR=1.028; p-value=0.005) and the rate of temperature changes (HR=1.841; p-value<0.001) are independent predictors of three-month all-cause mortality (primary outcome).

Interestingly, baseline glucose levels and its changes across hospitalisation were not predictors of poor clinical outcome at 3 months follow-up. One mentioned explanation was hyperglycaemia as a surrogate marker of critical illness, thus contributing to a higher mortality with no association to the poor clinical outcome.

Although this was not emphasised by the authors, another notable finding was the absence of an association between rise in blood sugar level and three-month mortality or poor clinical outcome, contrary to what other studies have shown.

Consistent with the literature, variation in BP values during hospitalisation was a predictor of all-cause mortality and poor functional outcome at three months after ischaemic stroke. Since cerebral autoregulation typically is impaired in the acute and sub-acute phase of stroke, it is likely that fluctuations in blood pressure compromise cerebral perfusion thus contributing to poor outcome.

Baseline temperature and slope of temperature change were both predictors of three-month all-cause mortality and poor clinical outcome. In addition to fever being also a surrogate index for nosocomial infections, the survival penumbra in the acute and sub-acute phase of ischaemic stroke can be influenced by changes in temperature, thus producing a worse functional outcome.

The variability between studies concerning parameters definitions, cut-off thresholds and statistical methods are important obstacles to interpretation and comparison. The large sample size, the assessment of serial values of parameters beyond the hyperacute phase and the careful adjustment for confounders are the strengths of this study.

Being a single centre study, with a long period of inclusion, makes the results extrapolation to other populations hazardous. Moreover, it is important to note that the etiology of the changes in glucose, BP and temperature were not discussed, which can also affect the mortality rate.

In conclusion, the results contribute to the understanding of of the complex physiology in acute ischaemic stroke. How to optimize supportive treatment of physiological factors such as blood pressure, blood glucose and temperature is yet to be answered. Hopefully, trials such as PRECIOUS (ISRCTN82217627) will bring us closer to a conclusion.