[Japan Coma Scale as a grading scale of subarachnoid hemorrhage: a way to determine the scale]

The grading scale for subarachnoid hemorrhage (SAH) with inter-grade outcome differences is essential for evaluating the effectiveness of newly developed therapeutic modalities. Although Hunt's grade and WFNS scale have been widely used, these grading scales do not meet this requirement. We previously proposed a revised WFNS scale based solely on the Glasgow Coma Scale (GCS) that has intergrade outcome differences of high-level significance. The Japan Coma Scale (JCS) has been long and widely used in Japan. The purpose of this study is to show whether it is possible to determine a reasonable SAH grading scale based on the JCS and to show a way to determine an SAH grading scale.

We retrospectively analyzed 1398 consecutive cases of aneurysmal SAH operated on within Day 7 of the latest onset. The preoperative JCS and GCS were evaluated just before the surgery and the Glasgow Outcome Scale (GOS), analyzed with numerical transformation (1 = dead to 5 = good recovery), was estimated at 6 months after the onset. All 510 possible combinations of scores of JCS were statistically tested under the following 2 assumptions; (1) JCS = 0 and JCS = 100 fall into a single independent grade. (2) No other single JCS score should fall into a single grade.

The outcome differences between JCS 0 and 1, and 100 and 200 are significant. The outcome difference between JCS 30 and 100 is relatively higher than any other set of 2 scores of JCS. Only 5 combinations are practical among the candidates to be analyzed. Out of 510 combinations, the following combination shows the highest inter-grade outcome differences; I (JCS = 0, n = 375, mean GOS = 4.78) II (JCS = 1, 2; n = 310; mean GOS = 4.47) III (JCS = 3-30; n = 476; mean GOS = 3.96) IV (JCS = 100; n = 96; mean GOS = 3.10) V (JCS = 200, 300; n = 141; mean GOS = 2.33). In JCS, the mean outcome of JCS = 3 is worse than those of JCS = 10, 20, and 30. The outcome difference between JCS 0 and 1 is only significant in patients over 60 years old.

Taking all the 510 possible combinations of JCS into consideration, we obtained a reasonable combination containing 5 grades. Although this grading scale showed good inter-grade outcome differences, JCS is not preferable to GCS as a consciousness evaluation system in the acute phase of SAH. We emphasize the importance of this way to determine a grading scale with a combinatorial approach, which can be applicable for re-evaluating the grading scales in the future.