Diagnostic value of neutrophil CD64, procalcitonin, and interleukin-6 in sepsis: A meta-analysis

Ke Wang, Shan Cong, Tiangang Ma, Xin Di, Chang Fu Tian, Min ZhaoResearch 2020 · DOI

This meta-analysis evaluates the diagnostic accuracy of three key biomarkers—neutrophil CD64 (nCD64), procalcitonin (PCT), and interleukin-6 (IL-6)—in identifying sepsis. By synthesizing data from 54 studies, the researchers found that nCD64 is the most reliable diagnostic tool overall, though PCT remains highly effective for assessing patients in critical condition.

TL;DR

Neutrophil CD64 (nCD64) outperforms procalcitonin (PCT) and interleukin-6 (IL-6) as a diagnostic biomarker for sepsis, boasting a superior Area Under the Curve (AUC) of 0.94. While nCD64 is the most accurate general "smoke detector" for infection, PCT remains a valuable tool for identifying sepsis in severe, high-stakes environments like the ICU.

What Problem Does This Paper Address?

Sepsis is a life-threatening medical emergency where the body’s response to infection causes systemic organ damage. Timing is everything: statistics show that if treated within the first hour, survival rates exceed 80%, but this drops to 30% after just six hours.

The "gold standard" for diagnosis—blood culture—is painfully slow (taking days) and prone to false negatives, especially if the patient has already started antibiotics. Clinicians need rapid, reliable biomarkers to distinguish between general inflammation and true sepsis. While PCT and IL-6 are commonly used, their accuracy varies, and the emerging marker nCD64 has shown promise but required a large-scale statistical validation to confirm its clinical value.

Methods

The authors performed a systematic meta-analysis of existing literature published up to December 2018.

  • Search Strategy: They queried major databases (PubMed, Medline, Web of Science, etc.) for studies comparing nCD64, PCT, and IL-6 in adult sepsis patients.
  • Quality Control: 54 high-quality articles were selected based on the QUADAS-2 (Quality Assessment for Diagnostic Accuracy Studies-2) tool.
  • Statistical Modeling: Using Stata 15.0 and MetaDisc 1.4, they calculated pooled sensitivity, specificity, and the Summary Receiver Operating Characteristic (SROC) curves. They employed random-effects models to account for the inherent "messiness" (heterogeneity) of clinical data across different hospitals and regions.

Datasets

The study did not use a single new dataset but rather pooled data from 54 independent studies involving a total of 9,842 participants.

  • nCD64: Evaluated in 20 studies.
  • PCT: Evaluated in 39 studies.
  • IL-6: Evaluated in 15 studies. The sepsis prevalence across these combined cohorts was 54.8%.

Key Insights and Ideas

  • The nCD64 Advantage: Neutrophil CD64 is a receptor that is barely present on "resting" white blood cells but spikes 10-fold within 4–6 hours of an infection. Unlike other markers, it remains stable in blood samples for a significant period, making it a robust diagnostic "trigger."
  • Severity Matters for PCT: While PCT was less accurate than nCD64 overall, it performed significantly better in ICU settings (AUC 0.86) than in non-ICU settings (AUC 0.82). This suggests PCT is a better indicator of the severity of the condition rather than just the presence of infection.
  • IL-6 is the "Early Bird" but Lacks Precision: IL-6 peaks very quickly (within 2 hours), but its overall diagnostic accuracy was the lowest of the three. It may be useful for very early screening but requires confirmation from other markers.
  • Regional Heterogeneity: The analysis revealed that nCD64 was particularly consistent and accurate in Chinese patient populations, whereas non-Chinese groups showed higher variability, likely due to differences in clinical settings and sample sizes.

Key Results

The diagnostic performance was ranked by the Area Under the Curve (AUC), where 1.0 is a perfect test:

Biomarker Sensitivity Specificity AUC
nCD64 0.88 0.88 0.94
PCT 0.82 0.78 0.87
IL-6 0.72 0.70 0.77

Fagan’s nomogram analysis showed that if a patient has a 50% pre-test probability of sepsis, a positive nCD64 test raises that probability to 88%, while a negative test drops it to 12%.

Limitations and Future Work

The authors identified several hurdles to immediate universal adoption:

  1. High Heterogeneity: There was significant variation between studies, likely caused by different "cut-off" values used by different labs to define a positive result.
  2. Publication Bias: There was evidence of publication bias specifically regarding nCD64 studies, which might slightly overstate its efficacy.
  3. Language Barrier: The search was limited to English and Chinese literature, potentially excluding relevant data from other regions.
  4. Need for Standardization: Future research must focus on establishing a universal "optimal cut-off value" for these biomarkers to ensure they can be used reliably in any hospital worldwide.

Summary updated

Abstract

Abstract The aim of the study was to conduct a meta-analysis to evaluate the accuracy of neutrophil CD64, procalcitonin (PCT), and interleukin-6 (IL-6) for the diagnosis of sepsis. The sample articles were searched in various databases to collect published studies on the diagnosis of sepsis by neutrophil CD64, PCT, and IL-6. By using the Stata SE 15.0 software, forest plots and the area under the summary receiver operating characteristic curves were drawn. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and area under the curve (AUC) were calculated. 54 articles were included in the study. The number of studies that evaluated the diagnostic value of neutrophil CD64, PCT, and interleukin-6 were 20, 39, and 15, respectively. The pooled sensitivity, specificity, and AUC of neutrophil CD64 for the diagnosis of sepsis were 0.88 [95% confidence interval (CI), 0.81–0.92], 0.88 (95% CI, 0.83–0.91), and 0.94 (95% CI, 0.91–0.96), respectively. The pooled sensitivity, specificity, and AUC of PCT for the diagnosis of sepsis were 0.82 (95% CI, 0.78–0.85), 0.78 (95% CI, 0.74–0.82), and 0.87 (95% CI, 0.83–0.89), respectively. Subgroup analysis showed that the AUC for PCT diagnosis of intensive care unit (ICU) sepsis was 0.86 (95% CI, 0.83–0.89) and the AUC for PCT diagnosis of non-ICU sepsis was 0.82 (95% CI, 0.78–0.85). The pooled sensitivity, specificity, and AUC of IL-6 for the diagnosis of sepsis were 0.72 (95% CI, 0.65–0.78), 0.70 (95% CI, 0.62–0.76), and 0.77 (95% CI, 0.73–0.80), respectively. Of the three biomarkers studied, neutrophil CD64 showed the highest diagnostic value for sepsis, followed by PCT, and IL-6. On the other hand, PCT showed a better diagnostic value for the diagnosis of sepsis in patients with severe conditions compared with that in patients with non-severe conditions.
BibTeX
@inproceedings{wang2020diagnostic,
  author = {Ke Wang and Shan Cong and Tiangang Ma and Xin Di and Chang Fu Tian and Min Zhao},
  title = {{Diagnostic value of neutrophil CD64, procalcitonin, and interleukin-6 in sepsis: A meta-analysis}},
  booktitle = {Research},
  year = {2020},
  doi = {10.21203/rs.3.rs-60957/v1},
}
Figure