Within an hour after the bodies arrive in their flag-draped coffins at Dover Air Force Base, they go through a process that has never been used on the dead from any other war.

Since 2004, every service man and woman killed in Iraq or Afghanistan has been given a CT scan, and since 2001, when the fighting began in Afghanistan, all have had autopsies, performed by pathologists in the armed forces medical examiner system. In previous wars, autopsies on people killed in combat were uncommon, and scans were never done.

The combined procedures have yielded a wealth of details about injuries from bullets, blasts, shrapnel and burns — information that has revealed deficiencies in body armor and vehicle shielding and led to improvements in helmets and medical equipment used on the battlefield.

The military world initially doubted the usefulness of scanning corpses but now eagerly seeks data from the scans, medical examiners say, noting that on a single day in April, they received six requests for information from the defense department and its contractors.

“We’ve created a huge database that’s never existed before,” said capt Craig T Mallak, 48, a Navy pathologist and lawyer who is chief of the armed forces medical examiner system, a division of the Armed Forces Institute of Pathology.

The medical examiners have scanned about 3,000 corpses, more than any other institution in the world, creating a minutely detailed and permanent three-dimensional record of combat injuries. Although the scans are sometimes called “virtual autopsies,” they do not replace old-fashioned autopsies. Rather, they add information and can help guide autopsies and speed them by showing pathologists where to look for bullets or shrapnel, and by revealing fractures and tissue damage so clearly that the need for lengthy dissection is sometimes eliminated. The examiners try to remove as many metal fragments as possible, because the pieces can yield information about enemy weapons.

One discovery led to an important change in the medical gear used to stabilise injured troops on the battlefield.

Col Howard T Harcke, a 71-year-old army reserve radiologist who delayed retirement to read CT scans at Dover, noticed something peculiar in late 2005. The emergency treatment for a collapsed lung involves inserting a needle and tube into the chest cavity to relieve pressure and allow the lung to reinflate. But in one case, colonel Harcke could see from a scan that the tube was too short to reach the chest

cavity. Then he saw another

case, and another, and half a dozen more.

In an interview, colonel Harcke said it was impossible to tell whether anyone had died because the tubes were too short; all had other severe injuries. But a collapsed lung can be life-threatening, so proper treatment is essential.

Colonel Harcke pulled 100 scans from the archives and used them to calculate the average thickness of the chest wall in American troops; he found that the standard tubing, five centimetres long, was too short for 50 per cent of the troops. If the tubing was lengthened to eight centimetres, it would be long enough for 99 per cent.

The findings were presented to the Army surgeon general, who in August 2006 ordered that the kits given to combat medics be changed to include only the longer tubing.

The medical examiners also discovered that troops were dying from wounds to the upper body that could have been prevented by body armor that covered more of the torso and shoulders. The information, which became public in 2006, led the military to scramble to ship more armor plates to Iraq.

The armed forces CT scanner, specially designed to scan entire corpses one after another, is the envy of medical examiners and crime laboratories around the country, and several states have asked captain Mallak and his colleagues for advice on setting up scanners.

Colonel Harcke said he hoped the technology would help to increase the autopsy rates at civilian hospitals, which now perform them only 5 per cent to 10 per cent of the time.

“We hope to return to a time where we were 50 years ago,” he said, “when autopsies were an important part of the medical model, and we continued to learn after death.”

—International Herald Tribune