Neuron-specific enolase for the differentiation between sepsis patients with or without encephalopathy in intensive care unit

To the Editor,

We have read with great interest the clinical study and subsequent commentary by Zhi et al. published in Systematic Reviews [1]. The study showed serum level of neuron-specific enolase (NSE) in sepsis-associated encephalopathy (SAE) patients was significantly higher than that in septic patients without encephalopathy. The higher the serum NSE level in SAE patients, the higher their mortality rate and incidence of adverse neurological outcomes. This holds promise as reliable serum diagnostic tools for the detection and longitudinal monitoring of SAE, which represents an intriguing and significant subject for exploration. However, further refinement of certain details could have increased the study’s depth and trustworthiness.

NSE is the predominant enolase isoenzyme predominantly expressed by neurons and neuroendocrine cells. Extensively investigated as a prognostic biomarker for neurological conditions, NSE possesses a serum half-life of approximately 24 h. Its quantification is routinely employed in clinical settings to evaluate the extent of neurological tissue damage. Elevated serum NSE levels, consequent to neuronal release in instances of neurological disorders or injuries, serve as a critical indicator for assessing neurological impairment [2]. Our research team has expressed significant concern regarding the role of NSE in neurocritical care, particularly within the context of intensive care unit (ICU) management of sepsis and its associated neurological complications [3,4,5,6]. In response, we have developed and registered a comprehensive protocol in PROSPERO (registration no. CRD42023398736) to assess the diagnostic and prognostic utility of NSE in patients with sepsis-associated encephalopathy (SAE). This protocol has been published in the Journal of Clinical Neuroscience [7].

First, there were different opinions on optimal effect measures of serum NSE levels by mean difference (MD) or standardized mean difference (SMD). When the original study outcomes employ consistent or similar measurement methods, the differences in outcome variables are minimal, and the units of study outcomes remain uniform; the MD was used. When the measurement methods, units, or means of each original study outcome vary significantly, substantial heterogeneity among outcome variables arises, preventing direct combination, and the SMD was used. Because serum sample collection time in ICU, assay methods, and design of clinical experiments in this meta-analysis were varied, so SMD should be used.

Second, the study reported that due to insufficient data disclosure, a sensitivity analysis could not be performed. To assess the robustness of the primary analyses, we employed a leave-one-out approach. This method revealed that the associations between serum NSE levels and severe adverse events (Fig. 1A), as well as between serum NSE levels and mortality outcomes in septic patients (Fig. 1B), remained consistent regardless of the exclusion of any single study. These results indicate that no individual study significantly influenced the overall outcomes, thereby affirming the statistical stability of the findings in this meta-analysis. All analysis were performed using STATA software (version 17.0, StataCorp, College Station, TX, USA) [6].

Fig. 1

Sensitivity analyses. A Sensitivity analyses for studies of serum NSE levels between SAE and NE patients. B Sensitivity analyses for studies of serum NSE levels between non-survival and survival patients

Third, two literatures included in this study, Ehler J. (2019) [8] and Orhun G. (2019) [9], only reported on septic patients without encephalopathy (NE) and did not provide a sample size for this group. Consequently, neither of these studies should be included in the meta-analysis comparing serum NSE levels between SAE and NE patients.

From a clinical perspective, serum neuron-specific enolase (NSE) levels provide critical insights for clinical decision-making, guiding therapeutic interventions and surveillance strategies. NSE has the potential to serve as a valuable biomarker in monitoring clinical conditions related to the diagnosis and prognosis of SAE. To elucidate the precise role of NSE in SAE, it is essential to achieve a thorough understanding of the pathophysiology of SAE, particularly in relation to the patient’s progression from the initial inflammatory response to the subsequent phase of sustained immunosuppression.

In summary, the study conducted by Zhi et al. endeavors to forecast the complications associated with SAE and the clinical prognosis for sepsis patients by analyzing serum NSE levels, thereby making a significant contribution to the advancement of the field. Nevertheless, further comprehensive and detailed research is imperative to fully elucidate the relationship between NSE and SAE.