In people who suffer multiple concussions or head trauma, the debilitating headaches, dizziness, and memory loss that follow can last three months or more. These symptoms, collectively known as post-concussion syndrome (PCS), can bench athletes permanently, and may lead to neurodegenerative disease. To what degree does ongoing neuronal damage contribute to PCS? A small study in the September 19 JAMA Neurology offers clues. Scientists led by Kaj Blennow, University of Gothenburg, Sweden, report that in the cerebrospinal fluid of hockey players who endure PCS the longest, neurofilament light protein (NfL) rises slightly, indicating their axons have been damaged. Moreover, some players with PCS have less Aβ in their spinal fluid, hinting that the peptide might have begun to accumulate in the brain.

"To our knowledge, this is the first study to show increased CSF levels of NfL and lower Aβ42 in some professional athletes with PCS due to repetitive mild traumatic brain injury,” wrote first author Pashtun Shahim of Sahlgrenska University Hospital, Mölndal, Sweden, to Alzforum. Shahim is now a postdoctoral associate at Washington University School of Medicine in St. Louis. “These biomarkers may be used to identify individuals with PCS at risk of developing progressive neurodegeneration.”

Blennow’s team has long sought biomarkers of traumatic brain injury (TBI) among athletes. Previously, they reported that in the days and weeks after a bout, NfL in the CSF of boxers rose and often stayed elevated (Zetterberg et al., 2006Neselius et al., 2012). They also found that total tau in the blood in the hours and days after a head trauma can help diagnose concussion, the most common type of mild traumatic brain injury, and help determine when an athlete can return to play (Mar 2014 news). For long-lasting symptoms such as those in PCS, it is unclear how big a role axonal injury plays. The authors were also curious to know whether PCS heightens the risk of developing a neurodegenerative condition called chronic traumatic encephalopathy (CTE). CTE is marked by neurofibrillary tangles and loss of neurons, as well as Aβ accumulation in about half of cases (Stein et al., 2015). It can develop in sports players, military personnel, survivors of physical abuse, and anyone who suffered some kind of mild repetitive brain injury.

To address these questions, Shahim and colleagues recruited 16 professional hockey players, average age 31, who reported PCS due to repeated, mild TBI (rmTBI). Seven of them recovered in less than a year and returned to play. Nine continued to experience typical PCS symptoms, including fatigue, sensitivity to light and sound, and insomnia, for more than a year, forcing them to retire. Each donated a CSF sample an average of four months after his last head injury. The researchers used those samples to measure NfL as a marker of axonal injury and the glial fibrillary acidic protein (GFAP) for astrogliosis. They also measured CSF Aβ and tau to look for signs of amyloid accumulation and tangle pathology in the brain, as well as neurogranin as a marker of synapse integrity. Each participant filled out the Rivermead Post-Concussion Symptoms Questionnaire (RPQ), a self-report of the severity of symptoms, when they donated the spinal fluid. The researchers compared player CSF against samples from 15 controls averaging 25 years old with no known history of head trauma.

NfL and Aβ stood out. CSF NfL was similar among controls and players as a whole, however, the authors found higher levels of the protein in players whose symptoms lasted more than a year. NfL level also correlated with higher RPQ score, meaning more severe symptoms, and with the number of concussions sustained over the person’s lifetime. NfL is a structural protein expressed in long, myelinated neurons of the subcortical white matter.

Aβ trended in the opposite direction, falling lower in the PCS group compared to the controls regardless of how long symptoms persisted. For total tau, phosphorylated tau, glial fibrillary acidic protein, and neurogranin, no difference emerged between groups.

Though the study was primarily cross-sectional, one athlete did agree to donate CSF four times over two years. This player was in his 30s and had PCS that persisted more than a year. In that patient, both NfL and neurogranin remained high.

The results suggest that rmTBI damages myelinated axons of the white matter, and can jump-start the process of amyloid accumulation, the authors wrote. Measuring levels of these two proteins could help tell how badly a player is injured, how likely it is that PCS will persist, and whether the player is at heightened risk for CTE, the authors claim. However, the results need to be validated in a larger study before incorporating them into a clinical routine, wrote Shahim.

Since total and phosphorylated tau in the player’s CSF matched control levels, it implies that rmTBI spares shorter, unmyelinated axons in the cortex, the authors suggested. This seems counter to the idea that PCS associates with later CTE, which is characterized by the accumulation of hyperphosphorylated tau and neurofibrillary tangles. However, tau could accumulate later or may not have reached levels high enough to reflect ongoing mild pathology, noted the authors.

Having objective ways to measure whether or not a player is injured severely enough to take a break from his or her sport would be valuable, said Jens Kuhle, University Hospital Basel, Switzerland. He said that the results need to be confirmed in larger cohorts, and noted that biomarkers measurable in blood serum will ultimately be more useful. He raised the caveat that NfL levels rise with age, and could confound the comparison between older athletes and younger controls in this study. He co-led a group of researchers who recently found that CSF and blood levels of NfL rise in β-amyloidosis, tauopathies, and certain synucleinopathies in mice and humans (Jun 2016 news). 

Suzanne Schindler of the University of Washington in St. Louis noted that the researchers only examined a small number of volunteers, and found a difference in NfL levels only in a post-hoc subset analysis. “This raises concern that the association between NfL and PCS is either very weak or not real,” she wrote to Alzforum in an email. Testing in a larger cohort will be important to validate the findings, she said.

This study offers a possible explanation for persistent PCS in some people, and examines biomarkers that may distinguish athletes at risk for future neurodegenerative changes, including CTE, wrote Robert Stern, Boston University School of Medicine, in an accompanying editorial. “If early detection of CTE was possible, perhaps while an athlete was still actively competing, recommendations for early retirement could be made and/or disease-modifying treatments could be initiated with the hopes of preventing future neurodegeneration and associated morbidity.” Stern is co-leading a large, multicenter effort to refine diagnostic criteria and discover validated biomarkers of CTE by studying former professional and college football players. He told Alzforum that he and his colleagues have enrolled the first handful of subjects and are in the early weeks of collecting initial data (Dec 2015 news).—Gwyneth Dickey Zakaib 


  1. Shahim et al. studied a fairly small sample (16 hockey players and 15 controls). There was no difference between CSF NfL levels in study participants with postconcussion syndrome versus controls. However, there was a slightly significant (p=0.04) difference in a subgroup analysis between study participants with postconcussion syndrome for more than one year versus participants with postconcussion syndrome for less than one year or versus controls. It is unclear whether this subgroup analysis was planned a priori. The lack of difference between CSF NfL in postconcussion syndrome versus controls, and the significant results only in a post hoc subset analysis, raises concern that the association between NfL and PCS is either very weak or not real. Aβ42 levels were lower in PCS versus controls, but the p value was borderline (p=0.05). Replication in a larger cohort will be important to validate Shahim et al.'s findings.

Make a Comment

To make a comment you must login or register.


News Citations

  1. For Hockey Players, Brain Damage Can Be Measured in Blood
  2. Blood NfL Looks Good as Progression and Outcome Marker
  3. Large Grant for Chronic Traumatic Encephalopathy Biomarker Study

Paper Citations

  1. . Neurochemical aftermath of amateur boxing. Arch Neurol. 2006 Sep;63(9):1277-80. PubMed.
  2. . Monitoring concussion in a knocked-out boxer by CSF biomarker analysis. Knee Surg Sports Traumatol Arthrosc. 2015 Sep;23(9):2536-9. Epub 2014 May 13 PubMed.
  3. . Beta-amyloid deposition in chronic traumatic encephalopathy. Acta Neuropathol. 2015 Jul;130(1):21-34. Epub 2015 May 6 PubMed.

Further Reading


  1. . Time course of cerebrospinal fluid inflammatory biomarkers and relationship to 6-month neurologic outcome in adult severe traumatic brain injury. Clin Neurol Neurosurg. 2016 Oct;149:1-5. Epub 2016 Jun 27 PubMed.
  2. . Fluid biomarkers for mild traumatic brain injury and related conditions. Nat Rev Neurol. 2016 Oct;12(10):563-74. Epub 2016 Sep 16 PubMed.

Primary Papers

  1. . Neurochemical Aftermath of Repetitive Mild Traumatic Brain Injury. JAMA Neurol. 2016 Nov 1;73(11):1308-1315. PubMed.
  2. . Cerebrospinal Fluid Biomarkers in Postconcussion Syndrome: Measuring Neuronal Injury and Distinguishing Individuals at Risk for Persistent Postconcussion Syndrome or Chronic Traumatic Encephalopathy. JAMA Neurol. 2016 Nov 1;73(11):1280-1282. PubMed.