Simple Prediction Rule for Walking After Spinal Cord Injury

From Medscape Medical News > Neurology

Megan Brooks

March 4, 2011 — Spinal cord injury experts have developed and validated a simple prediction rule to assess a patient's chances of walking independently 1 year after traumatic spinal cord injury.

The prediction rule combines the patient's age and the results of 4 neurologic tests and is more accurate than the American Spinal Injury Association/International Spinal Cord Society neurologic standard scale (AIS) grading system, the researchers say.

Joost J. van Middendorp, MD, from the Spine Unit, Radboud University Nijmegen Medical Centre, the Netherlands, and the European Multicenter Study on Human Spinal Cord Injury (EM-SCI) Study Group used ambulation outcome data from 492 patients enrolled in the EM-SCI to develop the prediction rule.

Their report appears in the March 4 online issue of The Lancet.

After logistic regression analysis, the final model included age (dichotomized at 65 years) and motor scores of the quadriceps femoris (L3) and gastrocsoleus (S1) muscles and light touch sensation of dermatomes L3 and S1.

Table. Clinical Prediction Rule Variables and Minimum and Maximum Scores
Variable Range of Test Scores Minimum Score Maximum score
Age ≥65 years 0 – 1 −10 0
Motor score L3 0 – 5 0 10
Motor score S1 0 – 5 0 10
Light touch score L3 0 – 2 0 10
Light touch score S1 0 – 2 0 10
Total −10 40

According to the investigators, the combination of age younger than 65 years and these 4 neurologic predictors showed "excellent discrimination" in distinguishing independent walkers at 1 year after injury from nonwalkers. The area under the receiver operating characteristics curve (AUC) was 0.956 (95% confidence interval [CI], 0.936 – 0.976; P < .0001).

A temporal validation study in a second group of 99 patients from Europe confirmed "excellent discriminating ability" of the prediction rule (AUC, 0.967; 95% CI, 0.939 – 0.995; P < .0001), the researchers report.

Patients with a score of −10 have a 0% chance of walking independently 1 year after injury, those with a score of 10 have about a 35% chance, those posting a score of 15 have about a 67% chance, and those with a score of 20 have about a 97% chance, the researchers say.

A post hoc analysis showed that the timing of examination (<24 hours, <72 hours, or <15 days after injury) had no significant effect on the accuracy of the prediction rule, they note.

Accuracy Tops AIS

Dr. van Middendorp told Medscape Medical News that the accuracy of the prediction rule was "significantly higher" (change in AUC, 0.058; 95% CI, 0.030 – 0.086; P < .0001) than was the accuracy of the AIS grading system (AUC, 0.898; 95% CI, 0.867 – 0.928; P < .0001).

"The prediction rule had a clear additional clinical value for the prediction of an individual's ability to walk independently in each of the AIS grades," he said. What's interesting, he added, was that the prediction rule had additional clinical value in particularly AIS grade B and C patients.

In AIS grade B, sensory but not motor function is preserved below the neurologic level and includes the sacral segments S4-S5. In AIS grade C, motor function is preserved below the neurologic level, and more than half of key muscles below the neurologic level have a muscle grade of less than 3.

"Particularly for these grades of severity, physicians experience difficulties in prognosticating," Dr. van Middendorp said.

Clinicians could use this rule to "counsel patients with traumatic spinal cord injury and their families during the initial phase after injury," the researchers write, and to set rehabilitation goals. It might also be helpful in stratifying patients in interventional trials.

The question remains, Dr. van Middendorp noted, "whether this rule is applicable in other settings/countries. To demonstrate this, external validation studies are required." He said his team is planning to conduct an external validation study in Australia. In practice, however, "physicians from all over the world may and probably will use the prediction rule without external validation," he speculated.

Accurate Prediction Possible

The authors of a linked Comment in The Lancet say this research suggests that light touch sensory testing is of similar prognostic value to pinprick sensory testing, "which is generally thought to be the most reliable prognostic indicator of neurological recovery.

"Indeed, these 2 modalities could be of similar prognostic value, in view of the fact that they are both transmitted in the spinothalamic tracts," note Professor Wagih Shafik El Masri and Dr. Naveen Kumar of Keele University and the Robert Jones and Agnes Hunt Orthopaedic Hospital, Oswestry, United Kingdom.

Dr. van Middendorp and colleagues report that although the addition of pinprick sensory result at L5 to their prediction rule resulted in a slightly higher AUC in the derivation group, its inclusion resulted in a marginally lower AUC in the validation group. Because they wanted the prediction rule to be as simple as possible, they included only the light touch sensory scores in the model, they explain.

Professor El Masri and Dr. Kumar say the researchers "should be congratulated for providing further strong evidence for the emerging view that accurate prediction of an individual's clinical ability to walk independently after traumatic spinal cord injury is possible."

They say further studies are needed to "assess the power of the various prognostic indicators and assess their value when applied at different times after injury."