Patients with severe ketoacidosis have altered cerebral autoregulation with hyperaemia pattern before treatment

Cerebral oedema (CE) is a severe complication in patients with diabetic ketoacidosis (DKA) with high morbidity and mortality.1 It is incidence is variable, being the symptomatic cases present between 0.3% and 1% of the patients admitted to the paediatric intensive care unit (PICU). The physiopathology of CE in patients with severe DKA is not fully understood.2 We hypothesised that patients with severe DKA have pretreatment alterations in cerebral blood flow that may predispose them to CE. This was a prospective observational study performed at a second-care level PICU from 2022 to 2024. All patients admitted to the PICU with a severe DKA were recruited. A control group of patients (<18 years, without neurological impairment) were recruited to compare the results (n = 14). Written parenteral consent was obtained before enrolment. The local Medical ethics review approved the study (PI-22-324). Once the patient was diagnosed with severe DKA was admitted to the PICU and treatment started according to the local PICU protocol based on the international guidelines.3 According to the local protocol, once the patient was diagnosed with DKA, continuous monitoring of rSO2 with the INVOS 7100 cerebral oximeter monitor System for rSO2 (Medtronic) was performed. The paediatric sensor was placed at the right patient's forehead. With this system, the values are recorded every 30 s and these data were used to calculate hourly mean rSO2 values. rSO2 values were considered to be abnormal if they were above 80% or below 55%, indicating cerebral hyperperfusion or hypoperfusion, respectively. The neuromonitoring was continued until patient PICU discharge. TCD ultrasonography (Philips CX 50 ultrasound) was performed with a 3–5 MHZ transducer. Both middle cerebral arteries (MCAs) were insonated. All infants were in a 30° head up position during TCD examination. Statistical analysis was performed using SPSS (R). Categorical variables were indicated as frequency (n) and percentage (%), whereas continuous variables were summarised as the median and interquartile range (IQR) because they did not follow a normal distribution. Continuous variables were compared using the non-parametric Kruskal–Wallis test. Probability values <0.05 were considered statistically significant. The graphical representation of the statistical results was carried out with the statistical program GraphPad®. The study included 28 patients: 14 patients with DKA and 14 healthy controls (Table S1). None of the DKA patients developed CE and all had GCS scores of 14 or 15 throughout the study period. All of them recovered completely without any neurological deficits at hospital discharge. All DKA patients except one had rSO2 values above the normal range (rSO2 > 80%) at admission (Figure 1). In the following 24 h, the evolution is towards normalisation of rSO2, while the patient receives treatment according to protocol. As shown in Figure 1, DKA patients presented higher values of rSO2 in comparison to the control group. TCD alterations were also observed on admission, with progressive improvement over time. Mean cerebral artery velocities at admission, 12 and 24 h were 130 cm/s (IQR 110–140), 120 cm/s (IQR 110–130) and 110 cm/s (IQR 90–120) respectively. These changes over time were statistically significant (p = 0.020). These velocities were higher than those observed in the control group, with statistically significant differences at hours 1 and 12, as shown in Figure 4. Pulsatility index results at admission, 12 and 24 h were: 0.55 (IQR 0.5–0.6), 0.53 (IQR 0.5–0.6) and 0.6 (0.5–0.65), with no statistically significant changes between these three-time points (p = 0.498). Differences in pulsatility index were observed between DKA patients and the control group, with lower values in patients with DKA, as depicted in Figure 1. No statistically significant correlations were observed between mean cerebral artery velocities and the rSO2. We have found that patients with DKA have a pattern of cerebral hyperaemia by elevated cerebral rSO2 and TCD with hyperaemia flow with low pulsatility index and high velocities before treatment DKA is started. Our results point out that DKA patient's present impaired cerebral regulation before treatment, through an increase in cerebral blood flow as shown by TCD results and cerebral hyperaemia up to vasogenic oedema as the high values of cerebral oximetry point, leading this population to the risk of CE. The fact that these findings are present before the symptoms of CE appear reveals that the increase of cerebral blood flow may be a therapeutic target to reduce the morbidity and mortality of CE and these patients should be closely monitored in the PICU in order to carry out a thorough neurological surveillance. Most studies reveal that signs and symptoms of CE appear within the first 12 h of treatment, but our results show that impaired cerebral autoregulation is affected for up to 24 h. Several factors may contribute to cerebral hyperperfusion and autoregulatory failure in children undergoing treatment for DKA.4 Classically it has been postulated that excessive fluid infusion and the decline in serum osmolality may play a role in DKA CE, but recent papers have shown that cerebral hyperaemia during DKA treatment is not affected by differences in fluid infusion rate, and nowadays is believed that the most accepted cause is the result of ischemia–reperfusion injury.5 This brain injury is a result of multiple combination of metabolic and hormonal factors, inflammation, loss of cerebrovascular autoregulation and reperfusion injury. The osmotic shifts and other therapy-related factors during treatment may increase cerebral injury in DKA patients. The high levels of rSO2 which reflect cerebral hyperaemia may be explained by two different pathways: One is the increase in PaCO2 after previous days of hypocapnia. Second is the high concentration of ketone bodies which affects the cerebral arterioles, decreasing the rate of cerebral oxygen consumption and altering the neuronal metabolism. Our study has an important limitation being the sample size small, which limits the generalisability of the findings. As a conclusion we have found that children with DKA have a pattern of cerebral hyperaemia as rSO2 and TCD values show before treatment and during the first 24 h. We believe these patients should be closely monitored in PICU due to their neurocritical condition. Javier Rodriguez-Fanjul: Conceptualization; funding acquisition; investigation; writing – original draft. Sara Bobillo-Perez: Conceptualization; investigation. Amaia Calvo Fernandez: investigation. Carlos Rodrigo Gonzalo de Liria: Supervision. This study was performed in its entirety at Hospital Germans Trias I Pujol. There was no financing for this research. The authors declare they have no conflict of interest. The authors disclose no potential financial or ethical conflicts of interest regarding the contents of this submission. The study was in accordance with the Helsinki Declaration and was approved by the Hospital Germans Trias I Pujol Ethical Investigational Committee. Table S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.