Reassessing accuracy of blood cell correction factor for traumatic lumbar puncture.

Objective: This study assessed the utility of correction formulas for procedural peripheral blood admixture with cerebrospinal fluid (CSF) in adult patients using in vitro addition of peripheral blood.

Methods: Participants from the Johns Hopkins CSF Disorders Center were recruited if greater than 18 years old, not acutely ill, and able to tolerate lumbar puncture (LP). Peripheral blood was added in vitro to acellular CSF and then cellular concentration and protein were quantified.

Results: Correction formulas for white blood cell (WBC) count determined by linear regression for samples with more than 100,000 RBCs per uL were inaccurate. Linear correction formulas close to 1 WBC per 1000 red blood cells (RBCs) were more precise than the formula based on peripheral blood WBC to RBC ratio. Correction formulas for protein level derived by linear regression had minimal error.

Discussion: This study confirms in a controlled manner that CSF admixture with peripheral blood cannot always be reliably corrected for in a predictable manner. Use of correction formulas for protein level and for WBC count when CSF admixture with peripheral blood has RBCs less than 10,000 per microliter is possibly useful but will need to be assessed in larger studies that can confirm diagnostic utility.