基于格拉斯哥昏迷评分(GCS)、脑干反射和自主呼吸评估，构建易用且更准确的量表，用于潜在脑死亡器官捐献(DBD)供者的早期识别。

回顾性纳入2021年1月1日至2023年6月30日中山大学附属第一医院收治并最终判定为脑死亡的昏迷状神经重症患者，作为回顾组(n＝175)。依据现有和本院既往研究，记录此类患者向脑死亡进展的相关参数，包含GCS、脑干反射(如瞳孔对光反射、角膜反射和前庭眼反射等)和自主呼吸状态。脑死亡判定依据中国现行的脑死亡判定标准，由本中心脑死亡判定专家完成。在回顾组中，筛选脑损伤发生后14 d内进展为脑死亡的相关因素，构建GSC-瞳孔对光反射-自主呼吸量表(以下简称GCS-P-R量表)。2023年7月1日至2024年1月31日期间，前瞻性纳入本中心收治并最终判定为脑死亡的昏迷状神经重症患者，作为前瞻组(n＝51)，验证GCS-P-R量表的预测效能。计量资料组间比较采用成组t检验，采用Logistic回归分析昏迷状神经重症患者在脑损伤后14 d内进展为脑死亡的危险因素。计算GCS-P-R量表和GCS判定昏迷状神经重症患者在脑损伤后14 d内进展为脑死亡的灵敏度和特异度，并采用χ²检验进行比较。P<0.05为差异具有统计学意义。

回顾组患者中，GCS＝3分(OR＝3.86，95%CI：1.57~18.21)，一侧瞳孔对光反射消失(OR＝6.83，95%CI：4.35~19.34)和自主呼吸频率<设定频率(OR＝4.16，95%CI：1.63~15.52)是昏迷状神经重症患者在脑损伤后14 d内进展为脑死亡的危险因素。GCS-P-R量表评分≤0可用于潜在DBD供者的早期识别。前瞻组患者中41例(80.4%)在脑损伤后14 d内进展为脑死亡。GCS-P-R量表评分≤0分和GSC≤6分预测在脑损伤后14 d内进展为脑死亡的灵敏度分别为92.7%和100%，特异度分别为43.5%和0，二者灵敏度差异有统计学意义(χ²＝4.898，P<0.05)，提示构建的GCS-P-R量表预测效能优于GCS。

引入瞳孔对光反射和自主呼吸评估的GCS-P-R量表，能够更准确地预测昏迷状神经重症患者在14 d内是否进展为脑死亡，可早期识别潜在DBD供者。

To early identify the potential brain death donors, a user-friendly and more accurate scale was constructed based on the Glasgow coma scale (GCS), brainstem reflex and spontaneous respiration assessment.

From January 1, 2021 to June 30, 2023, neurocritical patients in coma, admitted to the First Affiliated Hospital of SYSU and determined as brain death ultimately were retrospectively included as the retrospective group (n=175). Based on existing and previous studies of our center, the factors related to the progression to brain death of such patients were recorded, including GCS, brainstem reflexes (such as pupillary light reflex, corneal reflex, and vestibulo-ocular reflex), and spontaneous respiratory status. Brain death determination was according to the current criteria for brain death determination in China, and was completed by specialized doctors in our center. In the retrospective group, the relevant factors of the progression to brain death within 14 d after brain injury were screened, and were utilized to bulid the GCS-pupillary light reflex (P)-spontaneous respiratory (R) scale. From July 1, 2023 to January 31, 2024, the comatose neurocritical patients, admitted to our hospital and determined as brain death ultimately, were prospectively included as the prospectively group (n=51). In the prospectively group, the the efficacy of constructed GCS-P-R scale was verified. The comparison of quantitative data between groups was conducted with group t-test, and Logistic regression was used to analyze the risk factors for the progression of brain death in comatose neurocritical patients within 14 d after brain injury. The sensitivity and specificity of GCS-P-R scale and GCS were caculated in determining the progression to brain death in comatose neurocritical patients within 14 d after brain injury, and compared with chi square test. P<0.05 was indicated statistical significance of the difference.

In the retrospective group, statistical analysis showed that GCS=3 (OR=3.86, 95%CI: 1.57-18.21), disappearance of light reflex in unilateral pupil (OR=6.83, 95%CI: 4.35-19.34), and the frequency of spontaneous respiratory less than the set frequency (OR=4.16, 95%CI: 1.63-15.52) were the related factors with the progression to brain death within 14 d after brain injury. The GCS-P-R scale score≤0 can be used for early identification of potential organ donors after brain death. In the prospective group, 41 cases (80.4%) progressed to brain death within 14 d after brain injury. According to the GCS-P-R scale score ≤ 0, the sensitivity and specificity of predicting brain death progression within 14 d after brain injury were 92.7% and 43.5%, respectively. The sensitivity and specificity of predicting brain death within 14 d after brain injury based on GCS ≤ 6 were 100% and 0, respectively, the difference of sensitivity between the two groups was statistically significant (χ²=4.898, P<0.05), indicating that the constructed GCS-P-R scale has better predictive performance than GCS.

The GCS-P-R scale which includes pupillary light reflex and spontaneous respiratory assessment, can more accurately predict the progression to brain death within 14 d in comatose neurocritical patients, and can be used for early identification of potential organ donors after brain death.