Modeling treatment effect heterogeneity in prophylactic lumbar drainage: a Double Machine Learning reanalysis of EARLYDRAIN

Abstract

Aim: The EARLYDRAIN trial illustrated that prophylactic lumbar drainage (LD) could reduce poor outcomes in patients with aneurysmal subarachnoid hemorrhage, although not uniformly. We aim to reanalyze trial data using Double Machine Learning (DML) to estimate individualized treatment effects and identify patients or patient subgroups most likely to benefit.

Methods: We applied a DML framework with causal Random Forests to data from 287 patients randomized in the EARLYDRAIN trial to prophylactic LD or standard care. Six binary outcomes were analyzed: vasospasm, cerebral infarction, infection, favorable six-month outcome [modified Rankin Scale (mRS) ≤ 2], functional independence [Glasgow Outcomes Scale-Extended (GOS-E) ≥ 5], and shunt dependency. Average treatment effects (ATEs) and conditional ATEs (CATEs) were estimated. Uniform Manifold Approximation and Projection of the CATE values identified treatment-response clusters.

Results: Across the full cohort, prophylactic LD showed no consistent ATE across outcomes [e.g., mRS ATE: 0.02; 95% confidence interval (CI): -0.13-0.17]; CATE distributions revealed significant heterogeneity, with four treatment-response phenotypes. Younger patients with elevated intracranial pressure and lower drainage volumes derived greater benefit, and potentially reduced vasospasm risk. Patients older than 60 with higher systolic blood pressure, greater hemorrhage burden, and positive fluid balance experienced limited benefit and increased shunt dependency. A web-based application (https://earlydrain.streamlit.app) was developed to translate these findings into clinical decision support.

Conclusion: DML analysis of the EARLYDRAIN trial was able to substantiate the heterogeneity in effects of prophylactic LD from the initial trial. DML offers a scalable framework to reveal treatment heterogeneity masked in trial averages and support precision medicine in neurosurgery.

Keywords

Lumbar drainage / Double Machine Learning / early drain

References

[1] Wolf S,Barner C.EARLYDRAIN Study GroupEffectiveness of lumbar cerebrospinal fluid drain among patients with aneurysmal subarachnoid hemorrhage: a randomized clinical trial.JAMA Neurol2023;80:833-42 PMCID:PMC10277935

[2] Lee KS,Motiwala M,Arthur AS.Effectiveness of cerebrospinal fluid lumbar drainage among patients with aneurysmal subarachnoid hemorrhage: an updated systematic review and meta-analysis.World Neurosurg2024;183:246-253.e12

[3] Al-Tamimi YZ,Feltbower RG.Lumbar drainage of cerebrospinal fluid after aneurysmal subarachnoid hemorrhage: a prospective, randomized, controlled trial (LUMAS).Stroke2012;43:677-82

[4] Liang C,Guo S.Serial lumbar puncture reduces cerebrospinal fluid (CSF) infection during removal of hemorrhagic CSF in aneurysmal subarachnoid hemorrhage after endovascular coiling.J Biomed Res2018;32:305-10 PMCID:PMC6117608

[5] Machida A,Ishihara T.Syringomyelia due to lumbar spinal fluid drainage in the acute phase of subarachnoid hemorrhage: a case report.J Stroke Cerebrovasc Dis2018;27:e11-4

[6] Chen Q,Pei J.Effects of lumbar cerebrospinal fluid drainage on infection, rebleeding, clinical vasospasm, cerebral infarction, hydrocephalus and mortality among patients with aneurismal subarachnoid hemorrhage.Mol Cell Biomech2025;22:1056

[7] Chernozhukov V,Demirer M,Hansen C.Double/Debiased/Neyman Machine Learning of treatment effects.American Econ Rev2017;107:261-5

[8] Feuerriegel S,Melnychuk V.Causal machine learning for predicting treatment outcomes.Nat Med2024;30:958-68

[9] Kasuya H,Kagawa M.The effect of continuous drainage of cerebrospinal fluid in patients with subarachnoid hemorrhage: a retrospective analysis of 108 patients.Neurosurgery1991;28:56-9

[10] Bracard S.Vasospasm and delayed consequences.Interv Neuroradiol2008;14 Suppl 1:17-22 PMCID:PMC3328054

[11] Pavelka M,Ho J.Vasospasm risk following aneurysmal subarachnoid hemorrhage in older adults.J Neurosurg2023;139:1302-10 PMCID:PMC11122688

[12] Macdonald RL,Huo D.Factors associated with the development of vasospasm after planned surgical treatment of aneurysmal subarachnoid hemorrhage.J Neurosurg2003;99:644-52

[13] Regula JU,Ringleb PA.Cerebral vasospasm and delayed cerebral ischemia in intraventricular hemorrhage.Neurocrit Care2014;20:460-5

[14] Kissoon NR,Fugate JE,Wijdicks EF.Positive fluid balance is associated with poor outcomes in subarachnoid hemorrhage.J Stroke Cerebrovasc Dis2015;24:2245-51

[15] Xu S, Zheng B, Su B, et al. Efficient estimation of weighted cumulative treatment effects by double/debiased machine learning. arXiv 2023;arXiv:2305.02373v1. Available from https://www.semanticscholar.org/paper/Efficient-estimation-of-weighted-cumulative-effects-Xu-Zheng/753b15ce4fcbaa3a1806d9982e77aab4b63e0b7f [accessed 4 December 2025].