Followers of the game of hockey are known to rattle off statistics on goals scored and penalty minutes, but a new study of NHL concussions introduces a more sobering set of figures and charts to fans on the consequences when their hockey heroes take hits to the head.

The largest and most detailed analysis of concussions in the National Hockey League, published Monday in the Canadian Medical Association Journal, looked at seven regular seasons from 1997 to 2004.

It was based on physician reports from every team in the league and found a total of 559 concussions during regular season games, working out to a concussion rate of 5.8 for every 100 players, or an estimated 1.8 concussions per 1,000 player-hours.

The most common postconcussion symptom was headache, reported in 71 per cent of cases, followed by dizziness in 34 per cent, nausea in 24 per cent, neck pain in 23 per cent and blurred vision in 22 per cent. Amnesia was experienced in 20 per cent of players who had a concussion, and loss of consciousness in 18 per cent.

"We found four predictors of potentially more serious concussions, and that is postconcussion headache, postconcussion low-energy or fatigue, amnesia or a memory loss, and an abnormal neurologic exam," said lead author Dr. Brian Benson of the Sport Medicine Centre in the University of Calgary's faculty of kinesiology.

The findings illustrate that more education is needed for everyone involved in professional hockey about the potential adverse effects when players fail to report concussion symptoms after taking a hit or they keep playing with symptoms, Benson said in an interview.

He said the NHL is looking at the data closely.

"They've been careful not to make haphazard decisions and they certainly want to study this well. That takes time. Concussion changes from season to season. So we have longitudinal data over seven years where they can make informed decisions on any measures they want to take to try to reduce this injury."

The highest profile player to be sidelined by concussion in recent months is Pittsburgh Penguins captain Sidney Crosby, who has not played since Jan. 5. His injury will be included in the next round of data.

"The concussion working group right now is working on analyzing the data from 2006 to 2011, and certainly hope to report some of those findings within the year," said Benson.

In March, commissioner Gary Bettman announced new concussion protocols that involve players who display symptoms being given a test by a doctor in a quiet location before returning to the ice, rather than trainers doing an exam on the bench during a game.

As well, the league is working with the NHL Players' Association on equipment modifications.

In the study released Monday, time lost from the game ranged from 0 to 342 days. Of the 529 cases in which lost time was recorded, 31 per cent involved players missing more than 10 days of competition. In 11 per cent of these 529 cases of concussion, players continued to play and then later reported symptoms to medical staff after the game.

"We found some interesting trends -- one being a gradual increase in post-concussion time loss over the seven years of study," Benson observed. "That may be due to the concussions being severe or physicians being more conservative in their return-to-play decisions."

The highest concussion rate recorded was 7.7 per 100 players in the 2000-2001 regular season, while the lowest was 4.6 in 1997-98. The last year covered by the statistics -- 2003-04 -- saw a rate of 4.9 per 100 players.

In terms of player position, goalies were least susceptible to concussions -- sustaining just 4.5 per cent of the instances.

"Centremen appeared to be at more risk, two times more risk of concussion than wingers or defencemen," said Benson. "We didn't do statistical trends on this or analysis, but certainly it appears to be that centremen are at more risk of concussion."

Over the course of the study, almost 70 per cent of cases were first concussions, 22 per cent were a second concussion and six per cent were a third concussion. There was a smattering of players who sustained fourth and fifth concussions.

On average, time lost from the game increased 2.25 times for every recurrent concussion.

Benson said the study suggests that more should be done to teach coaches, trainers and medical staff to recognize and evaluate any suspected concussion.

"It also suggests that there potentially should be some more conservative or precautionary measures in the immediate postconcussion period particularly when an athlete experiences a postconcussion headache, low energy or fatigue, memory loss or amnesia, a recurrent concussion, and many different post-concussion symptoms."

Commenting on the study, Dr. Michael Cusimano, a neurosurgeon at St. Michael's Hospital in Toronto, said the study probably underestimates concussions because it covers a time period before there was widespread awareness about the dangers of concussion.

"There was probably some underrecognition and players were probably returning too early in some instances. There were a significant number that returned, in fact, to the game immediately. And that wouldn't be allowed by most doctors who know about the nervous system these days, and those players that return too early, not surprisingly, had significant complication rates."

Cusimano said the disclosure of data by the NHL is a move forward, but it's still not enough.

"What's lacking still, though, in all of this is a focus at primary prevention, that is preventing the brain injury before it occurs. The focus here has been to look at the brain injury after it occurs, and I think there's still a tremendous amount of work that we need to do, work on preventing them before they occur at all levels of play. And that's going to require significant leadership right from the NHL down."

An accompanying commentary in the CMAJ says complete physical and cognitive rest is recommended immediately after the injury and should continue until the symptoms resolve.

"The patient should then start to increase his or her level of exercise," wrote Dr. Peter Sojka of Mid Sweden University, Ostersund.

"The patient should start with light aerobic exercise, followed by light sport-specific exercise, then more complex and heavy training and, finally, return to full practice."

The journal also carries an analysis of concussion in youth ice hockey by Syd Johnson of Dalhousie University in Halifax, which noted that concussion is a common serious injury in youth ice hockey that affects up to 25 per cent of players per season by one estimate.

"Reducing rates of concussion requires eliminating bodychecking for all except elite hockey players aged 16 years and older, as per the recommendations of the Canadian Academy of Sports and Exercise Medicine," Johnson wrote.