Ⅰ．スポーツ医・科学サポートシステム

宝田 雄大

Ⅰ． 早稲田大学ラグビー蹴球部スポーツ医・科学サポートシステム

早稲田大学ラグビー蹴球部では、２００４年、競技力向上のためのスポーツ医・

科学サポートシステム（Sports Medicine & Science Support System: SMS system）

が立ち上がり、今年で５年が経とうとしている。

このシステムの主な特徴は、

① いわゆる“チームドクター制の廃止”とスポーツ医・科学サポート担当責任者

(Director of Sport Medicine & Science Support、以下略してSMSディレクター)

の配置

② セカンドオピニオン制の導入

③ 傷害別の専門医とのネットワークの構築

④ トレーナー、鍼灸師、栄養士、トレーニングコーチなどのスポーツ現場の専門

職の充実

⑤ 体力強化及びリハビリテーションプログラムの定量的な効果判定の徹底

などが挙げられる。

以下に、ラグビー部をモデルケースとしたスポーツ医・科学サポートの概要を

紹介する。またこのシステムとその活動が効果的であったか、有効であったか否

かは、究極的には、部の目的、“大学日本一”、を達成できたか否かにより決定さ

れると考える（もちろん、科学的に有効である、効果的であると思われる知見を提

供はするが、スポーツ現場では科学的に有効であること自体にあまり意味をなさ

ない。つまり、“スポーツを、身体運動を科学すること”、と“主に科学的な情報を

提供しその組織の目的達成の一助となること”とは本質的にそれらの目的が異な

る）。この意味において、過去4年間の3度の優勝（2004年度、2005年度、2007）

と準優勝（2006年度）の事実から、実施された取り組みが“有効的、効果的であっ

た”と解釈することに違和感を覚える読者は少ないのではないだろうか。なお、こ

1

れらはすでに宝田（２００８年）に報告されている内容を含んでいることを了承され

たい。

１． 概要

1995 年の国際ラグビー評議会によるオープン化宣言以後、職業ラグビー選手が誕

生した。その結果、選手の技術と体力レベルは飛躍的に向上し、それらの絶妙な共

同作業によって生み出されるラグビープレーは見る者に感動を与えている。しかしそ

の一方で、国内外の試合数の増加、十代の若い職業ラグビー選手の誕生、そしてオ

ープンラグビー推進のためのルール改正などにより、試合中の傷害発生率はオープ

ン化宣言前に比べ、プロ選手で約3倍、アマ選手で約2倍に激増している現状を無視

することはできない。このような選手を取り囲むラグビー環境の激変により、これまで

以上に選手は傷害予防と競技力向上のための体力増強を強いられていることは想

像に難くない。

そこで、より効果的な競技力向上のために、スポーツ医・科学サポートシステム

（Sports Medicine & Science Support System: SMS system）の確立を目指すべく、

2004 年 6 月、早稲田大学ラグビーオールド・ボーイズ（R.O.B.）倶楽部 故富永栄喜

（前会長、1960 年卒）、矢部達三（副会長、1966 卒）、及び志賀英一（前本部幹事長、

1960 年卒）の下、幹事会 技術・競技部会内（前部会長 矢部達三、1966 年卒）で、

中長期的な展望にたち、スポーツ医・科学関連事項を継続的に検討・協議していくこ

とが決定された。特に、主将として大学日本一にも輝かれた、前 OB 会長の富永氏と

故宝田雄弥氏（1960 年卒）との長い親交もあり、率直に部における診察、リハビリ、体

力強化などのスポーツ医科学関連事項の現状をご説明し、早急に、体系的なスポー

ツ医科学サポートの必要性などについてご相談させていただきやすかったこと、そし

てその必要性を良く理解しその立ち上げに尽力していただいたことは深く感謝してお

り、ここに改めて御礼を申し述べたい。

2

さて、部会長は技術・競技部会委員からスポーツ医・科学サポート担当責任者

(Director of Sport Medicine & Science Support、以下略してSMSディレクター)を任命し、

SMSディレクターはより効果的なSMS systemの確立及びその運営のために、早稲田

大学ラグビー蹴球部内のスポーツ医・科学分野に関連した取り組みと関係スタッフ間

の総合調整及びそれらの監査・指導を主導的におこなう。なお、SMSディレクターはこ

の取り組みの公平（中立）性及び専門性の観点から、現時点において以下の要件＊）

を満たすことが望ましい。但し、今後、これらの要件変更の可能性を全く否定するもの

ではないことを申し添えたい。

①早稲田大学 R.O.B.倶楽部会員であり、本部幹事会内に設置された技術・競技部

会員であること。

➁早稲田大学に常設された教育研究組織に所属する者で常勤の専任教員である

こと。

③博士の学位を有し、スポーツ医・科学において研究業績が認められること。

④プロ（国内トップリーグ含む）ラグビーチームに加え、他のプロスポーツ種目にお

いてプロ契約し、その指導実績が認められること。

＊ ）SMS systemの効果的な活用は、SMSディレクターの公平（中立）性及び高い

専門性に依存すると考える。したがって、SMS system内の機能上の特性を備

え発揮できる人材を選抜するために、その要件を慎重に検討しなければならな

い。また要件は、それぞれの組織や団体の諸事情あるいは、それらを取り巻く

環境により異なることは言うまでもない。

SMS system 内にはスポーツ医学セクションとスポーツ科学セクションの 2部門あり、

それらはそれぞれ <手術 ― リハビリテーション ― デイリーケア ― 応急処置

― カウンセリング ― 救急>と、<体力強化 ― 栄養管理・指導> の機能を有する

3

（図１）。スポーツ医学セクションでは、SMS ディレクター主動による、横断的な専門医

のネットワークの構築と専門医（R.O.B.倶楽部賛助医師）の選択・決定が、（医師の）

所属組織の力学に影響されることなく、（患者である選手あるいは当該ラグビーチー

ムに対して）より公正（中立）且つ効果的な医療サービスの提供を目指している。通常、

一人の、あるいは所属を同じくする医師（日本では主に整形外科医が関わっているが、

オーストラリアなどでは内科医が中心となっている）が、試合などに帯同し、選手の受

傷内容に関係なく、その医師あるいはその医師が所属する病院で対応する。この医

師あるいは医師集団をチームドクターと称している。

スポーツ科学セクションでは、ラグビー選手に必要と考えられる、体力の獲得・強化

に向け、研究者（博士号取得者でありスポーツ医学・科学における研究業績が認めら

れ、プロラグビーチームにおける現場指導経験を有する者）と現場指導者（各種のト

レーニングコーチや運動指導者など）との連携と協力が不可欠である。研究者には、

“ラグビー選手に必要な体力”について、科学（定量）的な手法により先行的な理解を

示すこと、そして各種トレーニングコーチや運動指導者には、その“ラグビー選手に必

要な体力”の強化に対する具体的な方法論の提案・選定とその厳格な実施、さらには

その効果の定量的な追跡を要望している。選手あるいはコーチ時代の経験則だけに

頼る偏った体力強化の指導は、理想とする“常勝早稲田ラグビー”の達成はおろか、

無意味な傷害の発生と選手育成の障害となる可能性も否定できない。

図１に示すように、SMS system は様々な職種のスタッフを有しており、その効果的

な活用は各スタッフ間の相互理解と協力があってはじめて実現する。スタッフ間の連

携を高めるために、メーリングリストによる情報の共有化、週一度のスタッフミーティン

グなどの取り組みを行っている。また、各スタッフの取り組みの質的向上を図るため

に、各スタッフには 1 年毎の委嘱による年更新を了解のうえで、年度ごとの活動報告

書提出を義務付けている。なお、SMS system における Medical Doctor (M.D.、医師免

4

許の有資格者)及び Doctor of Philosophy (Ph.D.、博士の学位取得者)を総じてチーム

ドクターと称する。また、試合などの応急処置の対応は医師にお願いしているが、特

にその医師をゲーム（グランド）ドクターと呼ぶ。

２．スポーツ医学セクション

① 診察

診察（手術含む）は、R.O.B.倶楽部賛助医師にお願いする。賛助医師は専門

医として、あるいは医学的な諸問題のアドバイザーとして協力していただける医

師で、脳、首、脊髄、肩、足首、膝などのラグビー選手に頻発する傷害部位ごと

に分かれている。受傷時のゲームドクターの応急処置や搬送先の病院での検

査結果を踏まえ、さらなる診察および手術実施先病院の選定を上述の傷害部

位別専門医所属病院を中心に、アスレティックリハビリテーション担当のメディカ

ルトレーナー（全米アスレティックトレーナーズ協会公認、NATA）と鍼灸師が監

督の承諾のもと、とりおこなう。選定後は、メディカルトレーナーを受傷選手に帯

同させ、受傷状況の通知や診察結果の正確な理解とその後の対応について、

専門的立場より選手をサポートする（図２）。また、セカンドオピニオン制の導入

により、複数の専門医の診察を受け、選手本人（両親含む）とメディカルトレー

ナー及び監督間で、最終的に依頼する専門医を決定する。もし意見が分かれ

た場合、選手本人の意思を最大限に尊重し、ディレクターが総合調整及び最終

決定をおこなう。

② リハビリテーション

リハビリテーションは、手術直後の基礎リハビリテーション（医療施設、病院

内でおこなうもの）とアスレティックリハビリテーション（主にトレーニングルーム

5

やグランドでおこなう）に分類され、これらの担当者はそれぞれ理学療法士

（PT）とメディカルトレーナーとする（図２）。但し、リハビリテーション全体の責任

者はメディカルトレーナーとし、練習復帰までをサポートする。なお、リハビリテ

ーション中の障害部位以外の体力強化については、アスレティックトレーナー

（日本体育協会公認）が担当する。競技復帰、練習に復帰するためには、受傷

部位における機能回復（メディカル測定）は基より、患部外の筋機能改善と受傷

前の体力レベルの維持または向上が求められる（体力測定）。

③ 応急（緊急）処置

練習試合、公式戦などにおける受傷時の応急処置は、ゲームドクターにお願

いする（図２）。その際、さらなる精密な検査が必要と考えられる傷害について

は、傷害部位別専門医所属病院を中心に、ゲームドクター、アスレティックリハ

ビリテーション担当のメディカルトレーナー（全米アスレティックトレーナーズ協会

公認、NATA）、および鍼灸師が監督の承諾のもと、依頼先を決定する。また、

菅平の長期にわたる合宿中の受傷後の流れを図３に示す。なお、練習や試合

など以外で急病人は、早稲田大学の契約健診機関でもある河北総合病院に搬

送する。

④ メディカルチェック

毎年４月初旬に、新入部員を対象として、実施する（図４）。その内容は、ラグ

ビー選手用問診表の記入に加え、血液・尿検査、心電図エコー、胸部レントゲ

ン、筋と関節の機能評価などである。筋と関節の機能評価は上井草ラグビー部

寮内で、それ以外は河北健診センター（杉並区）で、実施している。

6

⑤ デイリーケア

日々の練習や（練習）試合時のテーピング及び（ゲームドクター不在時の）応

急処置（鼻血や切り傷などの止血、脳震盪など）、と練習・試合後の疲労回復促

進のためのマッサージ、受傷直後の初期治療（一過的な痛み軽減を含む）など

は、鍼灸師やこれらの業務に対応可能な専門職（日本体育協会公認、アスレテ

ィックトレーナーなど）が担当する。

３．スポーツ科学セクション

① 栄養

栄養学の研究者（博士号取得者）の指導の下、部員の寮生を中心とした管

理・指導を管理栄養士が担当する。ラグビー競技の最大の特性は、激しいコン

タクトによる筋損傷である。したがって、多少なりとも、練習や試合後の早期回

復の一助となるような、より競技特性を活かした栄養管理と指導の再検討が必

要である。

② 体力強化

運動生理学及びトレーニング科学などの研究者によって示された知見

｛Takarada et al., 2002（参考資料１）; Takarada, 2003 （参考資料２）; Takarada

et al., 2006（参考資料 3）｝などをもとに、“ラグビー選手に必要と考えられる体

力”について、先行的な理解を示す（宝田、2002）ことは、より効果的な体力強

化には必要不可欠である。次に、この理解や指針に基づいて、具体的な運動、

強化プログラムの作成とその厳格な実施をおこなう。これらは、筋力トレーニン

グコーチ、フィットネスコーチ（主に持久力を中心としたグランドで実施する運動

指導者）、アスレティックトレーナー（傷害選手の患部外の体力強化）が担当す

7

る。おこなわれた様々なプログラムの効果判定は特異なケースを除き、月々に

実施される体力測定（図５）によりおこなわれ、測定結果（図６）はその後のプロ

グラムに反映されなければならない（図７）。

４．その他

これまでに紹介した、より直接的なサポートに加え、SMS system内で取り扱うべき

主な事項、２つを示す。

① 調査・研究

この目的は、「常勝早稲田ラグビー」を達成するための競技力向上である。調

査・研究される内容が高い可能性をもって、成果として具体的に選手やチームに

還元することが求められ、そう判断された場合に限ってその調査・研究実施を承

認する（図８）。

② 設備・物品・消耗品など

各種のトレーニング機器や測定装置、医療施設以外で使用可能な治療機器、

デイリーケアでの消耗品など SMS system に関連した設備、物品、消耗品の選定

と管理も、選手の競技力向上には欠かせない。たとえば、日々のグランドコンディ

ションも無視できず、ここ数年、（天然芝に比べ）人工芝での下肢の傷害が増加す

る傾向にある。案件ごとに、専門的な立場から、公正に判断し決定していかなけ

ればならない。

（参考文献）

1. 宝田雄大 早稲田大学ラグビー蹴球部におけるスポーツ医・科学サポート、スポー

ツ科学研究, 2008（投稿印刷中）

8

2. Takarada, Y., Y. Sato and N. Ishii. Effects of resistance exercise combined with

vascular occlusion on muscle function in athletes. Eur. J. Appl. Physiol. 86: 308-314,

2002.

3. Takarada, Y. Evaluation of muscle damage after a rugby match with special

reference to tackle plays. Br. J. Sports Med. 37; 416-419, 2003.

4. Takarada, Y., D. Nozaki and M. Taira. Force overestimation during

tourniquet-induced transient occlusion of the brachial artery and possible

underlying neural mechanisms. Neurosci Res, 54(1): 38-42, 2006.

5. 宝田雄大 「ラガーマンの肉体改造法」 ベースボール・マガジン社、東京 2002

9

SMS ディレクター

監督

R.O.B 倶楽部賛助医師 リハビリ 体力強化

（スポーツ医学） （スポーツ科学）

スポーツ医学（整形外科） スポーツ科学担当 （医師、ゲームドクター） （研究員）

図１スポーツ医・科学サポートシステム（Sports Medicine & Science Support System:

SMS system）組織図

アスレチックトレーナー(AT)

鍼灸師 筋力トレーニングコーチ

フィットネスコーチ

栄養 メディカルトレーナー

アドバイザー

理学療法士（PT）

10

診断の報告とその後

の方針の決定 初期リハビリテーションとアスレティックリハビリ

テーションの実地。

賛助医師・病院を中心に選定と決定

メディカルトレーナー帯同で診察

監督に報告

アスレティックリハビリテーション

定期診察又は定期カウンセリング＆測定

定期診察又は定期カウンセリング

メディカル測定・体力測定の実施

賛助医師又は専門医に

連絡・診察予約

グランドドクターによる疾患部位の確認

と応急処置内容の報告

初期リハビリテーション・治療

セカンドオペニオンへ

受傷（グランドドクターによる確認・応急処置）

他の病院へ移送

可

緊急

検査・診察の必要性あり

不可

選手本人を含め監督とミーティング

河北総合病院を希望

救急車で移送

手術・治療

図２受傷から競技復帰までの流れ

患部外トレーニング ストレングスコーチ・フィットネスコーチが指導

競技復帰

検査結果を撮影しＤに連絡

下山の必要性有り

緊急手術の必要性有り

検査結果を貸出し、Ｄの確認

下山の必要性無し

メディカルトレーナーが

各専門医への手配

治療・リハビリテーションを

実施

Ｄ・家族。監督に連絡、処置

の決定

通常の受傷後の流れに順ずる

受傷（応急処置）

メディカルトレーナーによる確認

ドクターチェック

検査の必要有り

Ｄ*

在中

検査の必要無し

Ｄ*

不在

救急

Ｄ*

在中

ＴＤ*

不在

家族・Ｄ・監督に随時連絡

真田クリニックへ搬送・検査（脳震盪の場合

は小林脳神経外科を受診）

*D：ゲームドクターの略

図３菅平合宿中の受傷後の流れ

入部説明会・問診表記入

検査結果

河北健診センターにて検査

（身体特性、血液・尿検査、心電図エコ

ー、腰部レントゲン）

トレーナーチェック

（既往歴、関節可動域、徒手抵抗テ

スト、関節不安定性などを中心に実

施）

練習参加

メディカルトレーナー（ＭＴ）が問診表を確認

監督に報告・相談**

通院

4 月 1 日 2・4・5･７･8 日 9 日以降 4 月下旬

名簿を河北健診センターに郵送

異常あり 異常なし

入部式

運動制限有り

問題なし

図４ ２００８年メディカルチェックの流れ

体力測定値記録用紙

～簡易型～ 実施日：200 年 月 日～ 場所： 天気：晴れ ・ 曇り 気温：約 度

●個人データ

名前（漢字） 名前（カタカナ） 生年月日 年齢 所属（学校）

住所（自宅） 電話（自宅） 携帯番号 E-mail ポジション

〒

★高めたい運動能力は何ですか？

1. 上半身の力強さ 2. 下半身の力強さ 3. 全身の爆発的なパワー発揮

4. ダッシュ力 5. 基礎的な持久力 6. 筋持久力

7. その他（ ）

●測定結果

①形態及び身体組成

身長（㎝） 体重（㎏） %fat LBM（㎏）

②筋力・パワー

a. 各種パワー測定

脚伸展パワー(W) 最大無酸素パワー(W) 垂直跳び（㎝） 立ち幅跳び（m）

（ ）

注）最大無酸素パワーの結果は、トレーニング値も（）内に記入すること。

b. 1RM（1 回反復可能重量）

ベンチプレス スクワット デットリフト 懸垂（最大回数）

1RM (kg)

③スピード（2 回試行し、良いタイムを採用）

30ｍダッシュ （秒） 50ｍダッシュ （秒） 100m ダッシュ （秒）

／ ／ ／

④筋持久力と全身持久力

400ｍ走 （秒） 1500ｍ走 （秒） 天候、気温など

図 5 体力測定記録表

早稲田大学ラグビー蹴球部 氏名 A チーム平均 ポジション平均 外人選手平均 U21ﾁｰﾑ平均 大学トップチーム 偏差値 偏差値

ﾌﾘｶﾞﾅ (チーム内） （過去2年間）

（詳細型+簡易型）体力測定結果一覧表 測定日 00/00/00 00/00/00 00/00/00 00/00/00 00/00/00 00/00/00

Yudai Takarada e-mail: 生年月日と年齢 27.8歳 27.2歳 29.9歳http://www9.plala.or.jp/yudai618/ ポジション WTB WTB-FB

形態計測 体格 身長(cm) 174.7（175.2） 181.2 178.6 186.4 178.6 174.9 39.8 41.0体重(kg) 81.8（82.2） 94.3 81.6 110.0 91.9 81.2 41.6 41.3

(右）上腕囲(cm) 33.2（31.2） 35.0 32.1 37.4 43.6 45.5（右）大腿囲(cm) 62.8（62.7） 64.2 60.3 69.1 47.2 47.0

頚囲(cm) 40（39.6） 42.9 40.4 45.6 38.4 40.0

身体組成 体脂肪率(%) 10.7（12.6） 14.5 11.0 14.9 57.5 58.9(空気置換法） 除脂肪体重(kg) 73（71.9） 80.1 72.5 94.0 42.8 43.4

無酸素性能力 筋力 ベンチプレス（kg) 99.1（力強さ） スクワット(kg) 141.0

懸垂（回） 11.4握力(kg) 60.3（61.8） 58.7 57.4 65.0 52.2 53.5

背筋力(kg) 198.0 187.4 178.1 56.9 56.1MVC（Nm)／S-MVC(Nm/cm2)

膝屈曲／伸展　比

パワー 垂直跳び(cm) 46.7 55.6 49.0立ち幅跳び(m) 2.52 2.69 2.56 2.39 2.36

スピード(100m走） 0～30m(秒） 4.390～50m(秒） 6.69

0～100m（秒） 11.25 12.62 11.67 12.92 13.10 66.7 70.5

乳酸性能力 400m走 （秒） 56.0 61.6 56.8 65.9 64.6 61.9 64.6（筋持久力） ミドルパワー (W)

有酸素性能力 1500m走 （秒） 345 361.8 349.3 331.9 340.4 48.6 48.0（全身持久力） 推定最大酸素摂取量 (ml/min/kg) 58.8 50.7 55.8 62.1 61.5

50.7 51.6

（総合体力評価値） フィットネススコア

図6　体力測定結果１

早稲田大学ラグビー蹴球部

★チーム内における体力特性とポジション内の測定値比較★

①チーム内の体力特性

②同一ポジション内の測定値比較

③測定結果の評価と今後のトレーニングの方向性

a. 測定結果より○総合体力（フィットネススコアより）： 非常に劣る・やや劣る・普通・やや優・非常に優○瞬発力やスピード（50m走と100m走タイムより）：　　　　　　　非常に劣る・やや劣る・普通・やや優・非常に優○筋持久力（400m走タイムより）：　　　　　　　　　　　　 非常に劣る・やや劣る・普通・やや優・非常に優○全身持久力（1500m走タイムより）： 非常に劣る・やや劣る・普通・やや優・非常に優

b. 今後のトレーニングの方向性レベル１：全身持久力の確保　→〔（220－年齢）X0.7〕以上の心拍数（分当たり）の走速度で30分以上走ってください。頻度は3回／週。　　　　　ポジション別目標1500m走タイムを目指して、トレーニングをおこなってください。クロスカントリーなども良いでしょう。

レベル2：筋持久力の向上　→20秒の運動時間と20~40秒の休息時間が設定されたインターバルトレーニングを20～30分、週3回おこなってください。　　　　　400m走の目標タイムは60秒です。

レベル3：基礎筋力の確保→筋持久力を維持しながら、基礎的な筋力を獲得しましょう。ベンチプレスとスクワット及びデットリフトの最大挙上重量(1RM)は、　　　　　それぞれ、体重の1.5~倍と2.2~倍であり、それらの重量を目標に筋力の増加を図りしましょう。決して、身体の見た目ではありません。中身が大切!

レベル4：急激なパワー発揮の獲得→筋持久力を維持しながら、獲得した筋力をより短時間で発揮できるよう、爆発的なパワー発揮を心がけましょう。　　　　　通常の筋力トレーニング以外に、ジャンプ系トレーニングやプライオメトリクスを取り入れましょう。その際、プライオメトリクステストを受けて下さい。　　　　　脚伸展パワー、垂直跳び、及び立ち幅跳びの目標値はそれぞれ、2500~w、60~cm、と2.7~mです。

レベル5：コンタクトフィットネスの向上→20秒の運動時間と20~40秒の休息時間が設定されたインターバルトレーニングの中で、高強度なコンタクト　　　　　（１０秒以内）を加味し、コンタクトフィットネスを向上させましょう。この成果は試合中のタックル数や走行距離によって確かめることができます。

その他：個別トレーニング相談の必要性あり→担当者に相談してください。Yudai Takarada

　図７体力測定結果２

0

20

40

60

80

無酸素性能力 乳酸性能力 有酸素性能力 チーム平均

A　選手の体力特性無酸素性能力

乳酸性能力

有酸素性能力

チーム平均

（筋持久力） （全身持久力）

偏差値

（瞬発力や筋力）

100-400-1500m平均速度の比較（ポジション：FB-WTB）

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

A B C D E F

ポジ

ション

平均

チー

ム平

均

目標

値

選手名

平均

速度

(m/s)

100m速度

400m速度

1500m速度

瞬発力や筋力

筋持久力

全身持久力

図 8 部員・施設を対象にした調査・研究の申請と承認まで

早稲田大学ラグビー蹴球部部員・施設を対象とする調査研究の届出について

承諾する

協定書の必要性

審査

早稲田大学ラグビー蹴球部部員・体制

対象研究計画書の提出

承諾しない

承諾する

早稲田大学ラグビー蹴球部部員を対象とした調

査・研究における協定書 面

接

施設利用あり

施設利用なし

早稲田大学ラグビー蹴球部施設利用願いと

早稲田大学ラグビー蹴球部施設利用許可書の提出

共同研究者又は担当者と調査・研究の実施

承諾しない

調査・研究結果の提出

内容の見直し・訂正

早稲田大学ラグビー蹴球部部員・体制対象研究

計画書（簡易型）の提出

簡易型

有り

無し

5.参考資料

①参考資料１

ORIGINAL ARTICLE

Yudai Takarada Æ Yoshiaki Sato Æ Naokata Ishii

Effects of resistance exercise combined with vascularocclusion on muscle function in athletes

Accepted: 23 August 2001 / Published online: 16 January 2002� Springer-Verlag 2002

Abstract The effects of resistance exercise combined withvascular occlusion on muscle function were investigatedin highly trained athletes. Elite rugby players (n=17) tookpart in an 8 week study of exercise training of the kneeextensor muscles, in which low-intensity [about 50% ofone repetition maximum] exercise combined with anocclusion pressure of about 200 mmHg (LIO, n=6), low-intensity exercise without the occlusion (LI, n=6), and noexercise training (untrained control, n=5) were included.The exercise in the LI group was of the same intensity andamount as in the LIO group. The LIO group showed asignificantly larger increase in isokinetic knee extensiontorque than that in the other two groups (P<0.05) at allthe velocities studied. On the other hand, no significantdifference was seen between LI and the control group. Inthe LIO group, the cross-sectional area of knee extensorsincreased significantly (P<0.01), suggesting that theincrease in knee extension strength was mainly caused bymuscle hypertrophy. The dynamic endurance of kneeextensors estimated from the decreases in mechanicalwork production and peak force after 50 repeated con-centric contractions was also improved after LIO,whereas no significant change was observed in the LI andcontrol groups. The results indicated that low-intensityresistance exercise causes, in almost fully trained athletes,increases in muscle size, strength and endurance, whencombined with vascular occlusion.

Keywords Athletes Æ Ischaemia Æ Muscle hypertrophy ÆMuscle endurance

Introduction

A number of studies have so far shown that enduranceexercise training (low resistance/high repetition training)induces in muscle an increase in endurance capacity,comprising increases in the total volume, number andsize of mitochondria, the activity of enzymes related tooxidative energy metabolism, capillary density, andglycogen content (Booth and Thomason 1991; Edstromand Grimby 1986; Holloszy and Booth 1976; Salmonsand Henriksson 1981; Saltin and Gollnick 1983). Inaddition, these adaptations of skeletal muscle toendurance training have been shown to be enhanced byischaemia (Sundberg 1994).

On the other hand, exercise of relatively high inten-sity and short duration performed during ischaemia maypromote improvements of anaerobic function in skeletalmuscle. For example, an increase in the numerical pro-portion of fast-twitch (type II) fibres has been demon-strated in the leg muscles of patients with heart failure,chronic obstructive lung disease, and peripheral vasculardiseases such as intermittent claudication (Hammarstenet al. 1980; Hilderbrand et al. 1991; Mancini et al. 1989).In addition, our recent study has shown that low-intensity [about 20% one repetition maximum (1RM)]resistance exercise combined with vascular occlusioncauses a transient, post-exercise increase in concentra-tion of plasma growth hormone as well as an elevatedelectrical activity of the muscle during exercise(Takarada et al. 2000a). When a study of such low-intensity exercise combined with vascular occlusion wasconducted for 16 weeks using the elbow flexors of oldwomen, it caused marked muscle hypertrophy andconcomitant increase in strength (Takarada et al.2000b). However, such a strong effect of low-intensityexercise combined with vascular occlusion may be spe-cific to subjects whose day-to-day physical activities areextremely low.

In the present study, we investigated the long-termeffects of low-intensity resistance exercise combined with

Eur J Appl Physiol (2002) 86: 308–314DOI 10.1007/s00421-001-0561-5

Y. Takarada (&) Æ N. IshiiDepartment of Life Sciences,Graduate School of Arts and Sciences,University of Tokyo, Komaba Tokyo 153-8902, JapanE-mail: CYM06016@nifty.ne.jpTel.: +81-3-54546870Fax: +81-3-54544317

Y. SatoSato Institute for Rehabilitation and Fitness,Tokyo 183, Japan

vascular occlusion in athletes having much higher levelsof physical activity than that in untrained subjects. Theresults showed large effects of low-intensity resistanceexercise combined with vascular occlusion in inducingincreases in muscle size and strength as well asimprovements of muscle endurance in high-intensityexercise.

Methods

Subjects

A group of 17 young male athletes (rugby players) volunteered forthe study. The subjects were divided into experimental [mean (SD)][age 25.9 (0.6) years, n=12] and untrained control [age25.4 (0.8) years, n=5] groups, and the former was further dividedinto groups undertaking training combined with vascular occlusion(n=6) and normal training (n=6) groups. Their physical charac-teristics are shown in Table 1. They had previously been engaged inresistance (weight-lifting) training for more than 5 years. All of thesubjects had previously been fully informed about the experimentprocedures to be used as well as the purpose of the study, and theirwritten informed consents were obtained. The study was approvedby the Ethics Committee for Human Experiments, University ofTokyo.

Regimes for exercise training

The exercises used were low-intensity with vascular occlusion (LIO)and low-intensity without vascular occlusion (LI) for the occlusivetraining and normal training groups, respectively. The subjects inthe experiment groups performed bilateral knee extension exercisein a seated position using an isotonic leg extension machine. Sincethey had been elite rugby players, we made them train both sidessimultaneously and equally, and the non-dominant side was usedfor measurements. The range of joint motion was from 0� to 90� (0�at full extension). In LIO, both sides of their thighs were trainedwith the proximal portions being compressed by a speciallydesigned elastic belt (width 33 mm, length 800 mm). The beltcontained a small pneumatic bag (width 25 mm, length 100 mm)along its inner surface, which could be connected to an electronicpressure gauge so as to monitor the occlusion pressure (modelM.P.S.-700, VINE Medical Instruments Inc., Japan). The meanocclusion pressure throughout the period of training was [mean(SEM)] 196.0 (5.7) mmHg.

The exercise was performed twice a week and lasted for 8 weeksincluding the period for instruction and orientation (16 sessions intotal). In each exercise session, the subjects in the training groupperformed four sets of exercise with an interset interval of 30 s. Theintensity of exercise was approximately 50% of the weight thatcould just be lifted once throughout the complete range of move-ment (1RM), which was determined in the initial stage of exercisetraining and kept unchanged throughout the period of training. Ineach set of LIO, the subjects repeated the lifting movement untilfailure, whereas in LI, they were instructed to match the number ofrepetitions performed by LIO. The mean repetition in each set was

16.3 (0.7). The total amounts of exercise in LIO and LI were cal-culated as load·total repetitions of the lifting movement, and were52,133.5 (6,038.8) kg·repetitions and 51,892.7 (5,868.6) kg·repeti-tions, respectively. The subjects were instructed to lift and lower theload at an approximately constant velocity, taking about 2 s foreach concentric and eccentric action. The vascular occlusion wasmaintained throughout the session of exercise which lasted forapproximately 10 min, and was released immediately after thesession of exercise. All of the exercise sessions were preceded by a10 min warm-up on a cycle ergometer at about 50% of the physicalexercise capacity and a stretching of the major muscle groups to betrained.

The subjects in the control group were not engaged in theexercise programme. They were instructed to maintain their normallevels of activity with no new exercise activity during the period ofthe experiment.

Measurements of muscle strength and endurance

Isokinetic torque-angular velocity relationships of knee extensorswere examined by using an isokinetic dynamometer (Myoret,Kawasaki Industry, Co. Ltd., Tokyo, Japan). The subjects werefamiliarized with the test procedure on several occasions prior tothe measurements. They sat on a chair with their backs upright andwith their left legs (non-dominant side) firmly attached to the leverof the dynamometer. A pivot point of the lever was accuratelyaligned with the rotation axis of the knee joint, and the requisiteaxial alignment of joint and dynamometer axes was maintainedduring the movement. Isokinetic strength was measured at presetangular velocities of 30, 90, and 180�Æs–1. The range of angularmovement of the knee joint was limited between 0 and 90� ofanatomical knee angle. The value of peak torque was measuredregardless of where it was developed within the range of movement.Three trials were made at each angular velocity, and the highestvalue obtained was used for further analyses. Isometric torque wasmeasured at a knee angle of 80�.

Dynamic endurance for knee extension was assessed before andafter the 8 week training period from recordings of 50 repeatedconcentric contractions (0.5 s contraction and 1.0 s rest) atan angular velocity of 180�Æs–1 on the isokinetic dynamometer(Thorstensson and Karlsson 1976). The mechanical work produc-tion was calculated by integrating force with respect to knee-jointexcursion by using a computer (Macintosh 8100/100AV). Themuscle endurance was estimated from the percentage decrease inboth the amount of mechanical work and the average value of peaktorque during the last ten contractions compared to those duringthe initial ten contractions.

Magnetic resonance imaging

To obtain cross-sectional images of the thigh, magnetic resonanceimaging (MRI) was performed by using a 0.5 T superconductingsystem (Gyroscan T5 II, Philips Medical Systems International,Best, The Netherlands) with a wraparound body coil. The coilcovered the whole thigh, including markers attached to the skin.Twelve serial sections were acquired with a 6–10 mm sectionalthickness and a 0.6–1.0 mm intersection gap. The field of view was350 mm. Pulse sequences for spin-echo T1-weighted images were

Table 1 Physical characteristics of subjects. Values are mean (SEM); n=6 for low-intensity exercise combined with vascular occlusion(LIO), n=6 for low-intensity exercise with no occlusion (LI), n=5 for untrained control group

LIO LI Untrained

Pre-training Post-training Pre-training Post-training Initial Final

Age (years) 25.3 (0.8) 26.5 (0.7) 25.4 (0.8)Height (cm) 179.3 (1.8) 179.8 (1.6) 181.0 (3.8) 180.5 (3.7) 180.2 (3.8) 179.8 (3.7)Body mass (kg) 88.9 (4.1) 89.1 (3.7) 92.4 (6.8) 91.5 (6.5) 91.5 (6.8) 92.8 (6.5)

309

performed with a repetition time of 500–552 ms and an echo timeof 20–25 ms. Two signal acquisitions were used. The scan matrixand reconstruction matrix were 205·256 and 256·256, respectively.The image acquisition was started immediately after the subjectswere placed in the supine position to minimize the effect of gravity-induced fluid shift. The time required for the whole sequence wasabout 4–6 min.

For each subject in the training group, the range of serial sec-tions was deliberately determined on longitudinal images along thefemur so as to obtain sections of identical portions before and afterthe period of exercise training. Among the photographs of the12 cross-sectional images obtained, those of two portions near themidpoint of the thigh were chosen for the measurements of musclecross-sectional area (CSA). Photographic negatives were digitizedinto an 8 bit grey scale at a space resolution of 144 pixels per inch,and stored in a computer using an Epson ART-8500G scanner.Determinations of tissue outlines and measurements of CSA formuscles and other tissues were made by using National Institute ofHealth Image (version 1.25) software. The measurements wererepeated three times for each image and their mean values wereused. Deviation in these three sets of measurement was less than2%. Measurements were made only for the LIO group.

Electromyogram

Electromyographic (EMG) signals were recorded from vastuslateralis muscle. Bipolar surface electrodes (5 mm in diameter) wereplaced over the belly of the muscle with a constant interelectrodedistance of 30 mm. The EMG signals were amplified, fed into afull-wave rectifier through both low (time constant, 0.03 s) andhigh (1 kHz) cut filters, and stored in a Macintosh 8100/100AVcomputer. Integrated EMG with respect to time (iEMG) was usedas an indicator of muscle-fibre recruitment during isometric torqueexertion (Bigland-Ritchie 1981).

Statistical analysis

All values are shown as mean (SEM). Because of the small n values,Wilcoxon signed ranks tests were used to compare differencesbetween pre- and post-training values within the same subjects. Toexamine differences between groups, one-way analyses of variance(ANOVA) with the Scheffe F-test post-hoc procedure was used.For all statistical analyses, the 0.05 level of significance was used.

Results

Changes in muscle strength following exercise training

Changes in force-velocity relationships after the 8 weektraining period are shown in Fig. 1. All values ofisokinetic torque were normalized to the pre-trainingvalues of isometric torque. The LIO showed significant

increases in isometric and isokinetic strengths at all thevelocities examined (Fig. 1A), whereas no change instrength was observed in the LI group (Fig. 1B) and

Fig. 1 Effects of exercise training on force-velocity relationships.Isokinetic torque-angular velocity relationships of knee extensormuscles were obtained before (unfilled circle) and after (filled circle)exercise training. All values of knee extension torque (P) werenormalized to the pre-training values of isometric (velocity=0)torque (Po), and means and SEM were plotted. A Exercise at lowintensity combined with vascular occlusion (LIO; n=6). B Exerciseat low-intensity with no vascular occlusion (LI; n=6). C Untrainedcontrol group (n=5). *Statistically significant changes frompretraining values within the same subjects (P<0.01). �Statisticallysignificant differences between LIO, LI and control groups(P<0.05)

c

310

untrained control group (Fig. 1C). When averaged overall velocities, percentage increases in strength after theexercise training were 14.3 (2.0)%, and 3.2 (2.3)% in theLIO and LI groups, respectively. The percentageincreases in isometric and isokinetic strengths after LIOwere significantly (P<0.05) larger than those in theother two groups at all velocities examined. Also, therewas no significant difference between LI and theuntrained control group.

Changes in muscle CSA following exercise trainingwith vascular occlusion

To see whether the increase in knee extension strengthafter LIO was mainly caused by muscle hypertrophy orneuromotor adaptation, MRI analysis was made onlyfor the LIO group. Typical examples of the CSA, MRIof identical, mid portions of the thigh are shown inFig. 2. These images were taken before (A) and after (B)the period of exercise training with vascular occlusion(8 weeks), and exhibit a marked increase (by approxi-mately 15%) in the CSA of knee extensors after theperiod of training. To reduce errors in measurementassociated with a slight mismatch between the sectionalportions obtained before and after the period of exercisetraining and incidental deformations of muscles duringthe processes of MRI, 2 sections around the mid portionof the femur, each separated by about 20 mm, wereselected from 12 serial sections, and mean tissue CSAwas obtained from these 2 sections. The LIO showedsignificant (P=0.002) increases in the CSA of kneeextensors compared to those before the exercise training(Fig. 3). The mean percentage increase in CSA of theextensors was 12.3 (0.8)%, whereas no significantchanges were observed in those of the knee flexors andfemur. The maximal isometric torque per unit CSA wasnot increased significantly after the training: pre-training3.8 (0.2), post-training 4.0 (0.2) NÆm–1. These resultssuggest that the increase in muscle strength after LIO(Fig. 1) is due primarily to the muscle hypertrophy.

Changes in muscle endurance following exercise training

Dynamic endurance for knee extension was assessedbefore and after the period of exercise training byrecording 50 repeated concentric contractions at anangular velocity of 180�Æs–1. The fatigue indices weredefined as the declines in the amounts of work and theaverage values of peak torque, comparing those in theinitial 10 contractions and those in the last 10 contrac-tions. Both indices in the LIO group decreased signifi-cantly from 63.7 (2.3) to 58.7 (2.3)% (P=0.002) andfrom 61.3 (2.1) to 53.7 (4.0)% (P=0.002), respectively(Fig. 4A). In contrast, no significant changes wereobserved in the LI group and untrained control group(Fig. 4B, C). The improvements in muscle endurancewere evaluated as percentage reductions in the fatigue

indices. Those for both the amount of work and thepeak torque were significantly (P<0.05) larger in LIOthan in the other two groups, whereas no significantdifference was observed between the LI and controlgroups. This indicates that the dynamic endurance ofknee extensors was improved only by LIO.

On the other hand, iEMG for the initial and last10 contractions showed no significant differencebetween pre- and post-training measurements in eitherthe LIO or control groups (Fig. 5). In the LIO group, nosignificant difference was observed between the declineof iEMG from the initial 10 to the last 10 contractions inthe pre-training measurements [36.2 (4.6)%] and that inthe post-training measurement [28.0 (15.8)%]. Althoughthe large deviations in both LIO and control groups mayhave hindered correct interpretations, these observationssuggested that, in the LIO group, the improvement ofdynamic endurance was primarily caused by metabolic

Fig. 2 Typical magnetic resonance images showing transversesections of the mid-thigh, taken before (A) and after (B) theexercise training combined with vascular occlusion lasting for8 weeks. The images show identical sections, mid-way along thefemur in the same subject. Bar indicates 60 mm

311

adaptations in the muscle fibres rather than an increasedresistance to fatigue in the nervous system.

Discussion

The present study showed that low-intensity resistanceexercise (approximately 50% 1RM) combined withvascular occlusion caused not only muscle hypertrophyand a concomitant increase in muscle strength, but alsoan improvement in muscle endurance. The major effectof LIO would have primarily been caused by theocclusion itself, as the LI made at the same intensity andamount showed no significant effect.

The subjects in the present study were highly trainedathletes, in which conventional resistance exercisetraining would not readily cause increases in muscle sizeand strength (Hakkinen et al. 1987). However, the CSA

Fig. 4 Percentage decreases in the amount of work and peaktorque during 50 repeated contractions. Those during the last10 contractions were compared with those during the initial10 contractions. Pre-(unfilled bars) and post- (filled bars) trainingvalues are shown as means and SEM. A Exercise at low intensitycombined with vascular occlusion (LIO; n=6). B Exercise at low-intensity with no occlusion (LI; n=6). C Untrained control group(n=5). *Statistically significant differences between pre- and post-training values within the same subjects (P<0.01). �Statisticallysignificant changes compared between LIO, LI and control groups(P<0.05)

c

Fig. 3 Cross-sectional areas (CSA) of knee extensor, flexor andfemur measured before (unfilled bars) and after (dotted bars)exercise training combined with vascular occlusion. Values areshown as means and SEM (n=6). *Statistically significant changeswithin the same subjects (P<0.01)

312

of knee extensor muscles and isokinetic knee extensionstrength increased significantly (P<0.01) after trainingfor a period of only 8 weeks (Fig. 1A, Fig. 3), suggestingthat the restriction of blood flow during the exerciseprovided effectively a new, not previously experiencedstimulus.

The externally applied compression restricted theblood circulation during the low-intensity exercise andthe resulting hypoxic and acidic intramuscular environ-ment would have induced additional motor-unitrecruitment to maintain the given level of force, therebyevoking an increase in the electrical activity of the

muscle (Takarada et al. 2000a), as has also been shownin contractions made in ischaemic (Moritani et al. 1992;Sundberg 1994) and fatiguing conditions (Miller et al.1996). Such an elevated activity of the muscle duringexercise would have been one of the factors involved inthe potent effect of the LIO in inducing muscle hyper-trophy.

Another factor to consider is hormone action.Kraemer et al. (1990) have demonstrated that a sufficientamount of high-intensity exercise (approximately 6 setsat an intensity of about 80% 1RM for large musclegroups) carried out with an interset interval as short as1 min transiently provokes more than a 100-foldincrease in the plasma concentration of growth hormone(GH). Since such a dramatic increase in plasma GHconcentration was not seen after exercise having a longerinterset interval (3 min), it has been speculated that localaccumulation of metabolites stimulates the hypophysealsecretion of GH. Our recent study with young malesubjects also showed that low-intensity (20% 1RM)exercise with vascular occlusion of the lower extremitiescaused a 290-fold increase in the plasma concentrationof GH, whereas no such effect was seen after the exercisewithout this occlusion (Takarada et al. 2000a). Thisstimulated secretion of GH may also play a part in thepresent effects of LIO.

One of the interesting findings in the present studywas the increase in the dynamic endurance of kneeextensors for 50 repeated concentric contractions(Fig. 4A), in spite of the increase in CSA of kneeextensors by 12.3% (Fig. 3). This increase in muscleendurance would have been primarily caused by anadaptation of muscle, e.g. increases in oxidative energymetabolism and acid-buffering capacity, rather than anincrease in the resistance to fatigue in the nervous sys-tem, because the exercise training did not cause anychange in the iEMG pattern in either the initial or thelast 10 of the 50 repeated contractions (Fig. 5).

High-intensity resistance training tends to cause inthe muscle a decrease in the aerobic capacity insteadof an increase in strength (Schantz 1982). Mitochon-drial density within a muscle fibre in elite powerliftershas been shown to be lower than that in untrainedsubjects (Tesch et al. 1984). On the other hand, thenumber of capillaries per unit muscle CSA in body-builders has been shown to be similar to that in un-trained subjects, with a substantial increase in thenumber of capillaries per muscle fibre (Schantz 1982;Tesch et al. 1984).

In the present study, low-intensity exercise trainingwas performed in combination with external compres-sion of the proximal portion of the thighs. Under suchconditions, both the venous outflow from and arterialinflow to the exercising muscle would have been con-siderably suppressed during the exercise, causing bothhypoxia and an accumulation of metabolites such aslactate. Both of these factors may have played parts inthe angiogenesis within the muscle. Recent studies haveshown that hypoxia is strongly related to the growth of

Fig. 5 Integrated electromyograms (iEMG) during the initial 10and the last 10 contractions during 50 repeated contractions,measured before (unfilled bars) and after (filled bars) the exercisetraining. All values of iEMG (means and SE) were normalized topre-training values of iEMG in the first 10 contractions. A Lowintensity exercise combined with vascular occlusion group, Buntrained control group

313

blood vessels at the developmental stage, although it isstill unclear whether the same mechanism operates inadults (Risau 1997).

As mentioned previously, type II fibres may berecruited preferentially or additionally during the exer-cise when the blood flow is suppressed (Moritani et al.1992; Sundberg 1994), so that more glycogen would beused as an energy source during LIO than during normalexercise at the same intensity and amount. The LIO maytherefore have induced an increase in the storage ofglycogen and an improvement of glycolytic capacity oftype II fibres (MacDougall et al. 1979, 1982). In addi-tion, a previous study using rats has reported thatintermittent stimuli of muscles combined with a restric-tion of blood flow caused increases not only in muscleglycogen content but also in protein content (Elanderet al. 1985). Therefore, the present LIO may have beeneffective in increasing the glycolytic capacity of type IIfibres through increases in glycogen content and glyco-lytic enzyme activity.

In conclusion, low-intensity resistance exercise com-bined with vascular occlusion caused, in almost fullytrained athletes, increases in muscle size, strength andendurance. Neural, hormonal and metabolic factorswould have been involved in these combined effects.

References

Bigland-Ritchie B (1981) Relations and fatigue of human voluntarycontractions. In: Milner DI (ed) Exercise and sport sciencesreviews, vol 9. Franklin Institute Press, Philadelphia, Pa. pp75–117

Booth FW, Thomason DB (1991) Molecular and cellular adapta-tion of muscle in response to exercise: perspectives of variousmodels. Physiol Rev 71:541–585

Edstrom L, Grimby L (1986) Effect of exercise on the motor unit.Muscle Nerve 9:104–126

Elander A, Idstrom J-P, Holm S, Schersten T, Bylund-Fellenius A-C(1985) Metabolic adaptation to reduced muscle blood flow. II.Mechanisms and beneficial effects. Am J Physiol 249:E70–E76

Hakkinen K, Komi PV, Alen M, Kauhanen H (1987) EMG,muscle fiber and force production characteristics during 1 yeartraining period in elite weight-lifters. Eur J Appl Physiol56:419–427

Hammarsten J, Bylund-Fellenius A-C, Holm J, Schersten T,Krotkiewski M (1980) Capillary supply and muscle fibre types

in patients with intermittent claudication: relationships betweenmorphology and metabolism. Eur J Clin Invest 10:301–305

Hilderbrand IL, Sylven C, Esbjornsson M, Hellstrom K, Jansson E(1991) Does chronic hypoxaemia induce tranformations of fibretypes? Acta Physiol Scand 141:435–439

Holloszy JO, Booth FW (1976) Biochemical adaptations toendurance exercise in muscle. Annu Rev Physiol 18:273–291

Kraemer WJ, Marchitelli L, Gordon SE, Harman E, Dziados JE,Mello R, Frykman P, McCurry D, Fleck SJ (1990) Hormonaland growth factor responses to heavy resistance exercise pro-tocols. J Appl Physiol 69:1442–1450

MacDougall JD, Sale DG, Moroz JR, Elder GCB, Sutton JR,Howald H (1979) Mitochondrial volume density in humanskeletal muscle following heavy resistance training. Med SciSports 11:164–166

MacDougall JD, Sale DG, Elder GCB, Sutton JR (1982) Muscleultrastructural characteristics of elite powerlifters and body-builders. Eur J Appl Physiol 48:117–126

Mancini DM, Coyle E, Coggan A, Beltz J, Ferraro N, Montain S,Wilson JR (1989) Contribution of intrinsic skeletal musclechanges to 31PNMR skeletal muscle metabolic abnormalities inpatients with chronic heart failure. Circulation 80:1338–1346

Miller KJ, Garland SJ, Ivanova T, Ohtsuki T (1996) Motor-unitbehavior in humans during fatiguing arm movements. J Neu-rophysiol 75:1629–1636

Moritani T, Michael-Sherman W, Shibata M, Matsumoto T,Shinohara M (1992) Oxygen availability and motor unit activityin humans. Eur J Appl Physiol 64:552–556

Salmons S, Henriksson J (1981) The adaptive response of skeletalmuscle to increased use. Muscle Nerve 4:94–105

Saltin B, Gollnick PD (1983) Skeletal muscle adaptability: signifi-cance for metabolism and performance. In: Peachey LD,Adrian RH, Geiger SR (eds) Handbook of physiology, skeletalmuscle. American Physiological Society, Bethesda, Md., pp555–631

Schantz P (1982) Capillary supply in hypertrophied human skeletalmuscle. Acta Physiol Scand 114:635–637

Sundberg CJ (1994) Exercise and training during graded legischaemia in healthy man with special reference to effects onskeletal muscle. Acta Physiol Scand [Suppl] 615:1–50

Takarada Y, Nakamura Y, Aruga S, Onda T, Miyazaki S, Ishii N(2000a) Rapid increase in plasma growth hormone after low-intensity resistance exercise with vascular occlusion. J ApplPhysiol 88:61–65

Takarada Y, Takazawa H, Sato Y, Takebayashi S, Tanaka Y, IshiiN (2000b) Effects of resistance exercise combined with moder-ate vascular occlusion on muscular function in humans. J ApplPhysiol 88:2097–2106

Tesch PA, Thorsson A, Kaiser P (1984) Muscle capillary supplyand fiber type characteristics in weight and power lifters. J ApplPhysiol 56:35–38

Thorstensson A, Karlsson J (1976) Fatiguability and fibre compo-sition of human skeletal muscle. Acta Physiol Scand 98:318–322

Risau W (1997) Mechanisms of angiogenesis. Nature 386:17, 671–674

314

②参考資料２

doi:10.1136/bjsm.37.5.416 2003;37;416-419 Br. J. Sports Med.

  Y Takarada  

with special reference to tackle playsEvaluation of muscle damage after a rugby match

http://bjsm.bmjjournals.com/cgi/content/full/37/5/416Updated information and services can be found at:

These include:

References

  http://bjsm.bmjjournals.com/cgi/content/full/37/5/416#BIBL

This article cites 32 articles, 8 of which can be accessed free at:

Rapid responses http://bjsm.bmjjournals.com/cgi/eletter-submit/37/5/416

You can respond to this article at:

serviceEmail alerting

top right corner of the article Receive free email alerts when new articles cite this article - sign up in the box at the

Topic collections

(1100 articles) Sports Medicine � (1000 articles) Molecular Medicine �

  Articles on similar topics can be found in the following collections

Notes  

http://www.bmjjournals.com/cgi/reprintformTo order reprints of this article go to:

http://www.bmjjournals.com/subscriptions/ go to: British Journal of Sports MedicineTo subscribe to

on 15 December 2005 bjsm.bmjjournals.comDownloaded from

ORIGINAL ARTICLE

Evaluation of muscle damage after a rugby match withspecial reference to tackle playsY Takarada. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Br J Sports Med 2003;37:416–419

Objective: To investigate blood indices of muscle damage after a competitive rugby match.Methods: Fifteen elite amateur rugby players volunteered to participate (mean (SE) age 26.6 (0.7)years, height 179.8 (1.0) cm, weight 87.4 (2.2) kg, and VO2MAX 58.5 (1.2) ml/kg/min). The study wasconducted after two competitive matches during the 1999–2000 season. Plasma concentrations of lac-tate, potassium (K+), sodium (Na+), and myoglobin, and the activity of creatine kinase were measuredbefore and after the matches. In addition, the number of tackles by and on each subject and the aver-age duration of the work and rest periods were analysed using video recordings of the matches.Results: Myoglobin concentration and creatine kinase activity showed appreciable transient increasesafter the match. Peak values for myoglobin concentration (980 (166) µg/l) and creatine kinase activ-ity (1081 (159) U/l) were observed 45 minutes and 24 hours after the match respectively. Positive andsignificant correlations were observed between the number of tackles and both peak myoglobin con-centration (r = 0.85, p<0.01; n = 14) and peak creatine kinase activity (r = 0.92, p<0.01; n = 14).Plasma lactate and K+ concentrations also showed appreciable increases after the match, whereasplasma Na+ concentration showed a gradual decrease. The mean duration of the work and rest peri-ods were 21.5 (2.2) and 24.3 (3.1) seconds respectively.Conclusions: The rugby matches resulted in serious structural damage to the muscles, the extent ofwhich was highly dependent on the number of tackles.

Exercise induced muscle damage has been widely studiedafter various types of exercise in humans.1–6 Muscle dam-age after exercise results in a substantial increase in myo-

cellular protein levels in the blood.7 Creatine kinase in theblood is often used as an indirect indicator of muscle damage,its release from muscle tissue into the blood being associatedwith disruption of the muscle cell membrane.8–10 The extent ofmuscle damage has been related to both the intensity andduration of exercise, with intensity playing the main role.11 Forexample, the peak activity of creatine kinase after a marathonrace ranges from 500 to 3300 U/l,5 12 whereas its peak activityafter exercise with high force eccentric muscle contraction canreach 25 000 U/l.13 Such severe muscle damage immediatelyresults in a reduction in muscle strength of more than 50%,which is completely restored after about 10 days.4 14

Exercise induced muscle damage is mainly caused by strongmuscle contraction during intense exercise. In addition to thisendogenous factor, exogenous factors are considered to be animportant. Direct excessive muscle damage occasionally leadsto trauma graded as an injury. In multiple sprint sports suchas soccer, rugby, and field hockey, muscle is often damaged bydirect impact during exercise in which players contact or col-lide with each other, even without accompanying exogenoustraumas. Zuliani et al15 reported that a normal boxing match ofthree three minute rounds with two intervals of one minute(real boxing) results in large increases in the serumconcentration of creatine kinase and myoglobin, whereas thesame exercises without direct contact (shadow boxing) donot, suggesting that the muscle damage observed after realboxing is mainly caused by direct punches to the body. Therehave been few studies on muscle damage after competitivematches in multiple sprint sports involving body contact suchas soccer, rugby, and field hockey in which the relationbetween exercise with body contact and muscle damage hasnot been mentioned.

This study investigated muscle damage in rugby players

after competitive matches involving fierce body contact.

METHODSSubjectsFifteen elite Japanese amateur rugby union players aged

23–30 volunteered to participate. Their physical characteristics

were: mean (SE) height, 179.8 (1.0) cm; body mass, 87.4 (2.2)

kg; maximal oxygen consumption (VO2MAX) on a treadmill,

59.5 (1.2) ml/kg/min. Playing positions included two props,

one hooker, two locks, three loose forwards, three centres, two

wings, and two fullbacks. Written informed consent was

obtained from each subject. The study was approved by the

ethics committee for human experiments, University of Tokyo.

Experimental proceduresThe study was conducted after two competitive matches in the

East Japan Company Rugby Football League during the 1999–

2000 season. The matches took place on 14 November and 28

November with no competitive matches between. The matches

were played in good weather, and the outdoor temperature

was 8–15°C. Plasma creatine kinase activity and plasma

concentrations of myoglobin, lactate, potassium (K+), and

sodium (Na+) were measured before and after each match.

Blood samplingVenous blood samples (20 ml for each point of measurement)

were obtained through an indwelling cannular in a superficial

arm vein from the subjects seated in a slightly reclined

position. All blood sampling was conducted at the same time

of day (1200–1400 hours) to reduce effects of diurnal

variation. A resting blood sample was obtained after the sub-

jects had sat quietly in a slightly reclined position for 20 min-

utes 48 hours before each match. After each match, blood

samples were obtained at 0 (immediately after the match), 45,

and 90 minutes and at 24, 48, and 72 hours. All blood samples

were processed and stored at −20°C until analysis. The subjects

were asked to refrain from ingesting alcohol and caffeine and

performing any strenuous exercise for 48 hours before and 72

hours after each match.

. . . . . . . . . . . . . . . . . . . . . . .Y Takarada, School ofSport Sciences, WasedaUniversity, Saitama359-1192, Japan

Correspondence to:Dr Takarada, Millenium 02# 201, 2-20-7Kamishakujii, Nerima-ku,Tokyo 177-0044, Japan;cym06016@sea.plala.or.jp

Accepted 25 September2002. . . . . . . . . . . . . . . . . . . . . . .

416

www.bjsportmed.com

on 15 December 2005 bjsm.bmjjournals.comDownloaded from

Biochemical analysisPlasma concentrations of K+ and Na+ were determined by

potentiometry (model 7450 autoanalyzer; Hitachi, Tokyo,

Japan) with ion selective electrodes (Daiichi Pure Chemicals,

Tokyo, Japan). Plasma concentrations of lactate and myo-

globin were measured by spectrophotometry using a lactate

dehydrogenase coupled enzymic system16 and radioimmuno-

assay using a test kit (Daiichi Radioisotope Laboratory Ltd,

Tokyo, Japan) respectively. Plasma creatine kinase activity was

determined by spectrophotometrically measuring NADPH

formed by hexokinase and the D-glucose-6-phosphate dehy-

drogenase coupled enzymic system.

Match analysisThe number of tackles in which each subject was involved and

the mean duration of the work and rest periods were investi-

gated by analysing video recordings of the two matches. The

number of tackles was defined as the total number of times

that the player tackled or was tackled in situations in which

the player was tackled from in front. The durations of the work

and rest periods were defined respectively as the period from

the beginning of a play to the interruption of that play by the

referee, and the period from the interruption of a play by the

referee to the start of the next play.

Statistical analysisUnless otherwise stated, variables are expressed as mean (SE).

The Wilcoxon signed ranks test was used to compare

differences between variables measured before and after the

match within the same subjects. Correlation analysis was car-

ried out by linear regression, and the Pearson correlation coef-

ficient (r) was calculated. For all statistical analyses, p<0.05

was considered significant.

RESULTSOne subject was badly bruised on the right thigh and dropped

out of the match on 14 November 15 minutes into the second

half. The remaining subjects completed the match (80

minutes). Therefore all variables are expressed as mean (SE)

for the remaining 14 subjects.

Match analysesThe mean (SD) number of tackles was 14.0 (7.4) per match.

The relation between the number of tackles and any muscle

damage after the match was investigated. The mean (SD)

duration of the work and rest periods in the two matches was

21.5 (2.2) and 24.3 (3.1) seconds respectively.

Myoglobin concentration and creatine kinase activity inthe plasmaFigure 1 shows plasma myoglobin concentration and plasma

creatine kinase activity measured before and after the two

matches. Both had significantly increased after the matches. A

peak was reached 45 minutes after the match for myoglobin

and 24 hours after the match for creatine kinase. They then

returned rapidly to their resting levels in an exponential fash-

ion. Peak myoglobin concentration and peak creatine kinase

activity in the subject with a bruised right thigh resulting from

a tackle by an opposing player were 1641 µg/l (45 minutes after

the match) and 2621 U/l (24 hours after the match)

respectively. These values are much higher than in the remain-

ing 14 subjects (myoglobin, 980 (166) µg/l; creatine kinase,

1081 (159) U/l). In addition, creatine kinase activity 48 and 72

hours after the matches was 1620 U/l and 1025 U/l respectively

and remained at a much higher level than in the other subjects.

The relation between the number of tackles and muscle

damage after the match was investigated. Figure 2 shows the

relation between the number of tackles and peak myoglobin

concentration (45 minutes after the match) and peak creatine

kinase activity (24 hours after the match). Positive and

significant correlations were observed between both. These

results suggest that the extent of muscle damage depends on

the number of tackles.

Plasma concentrations of lactate, K+, and Na+

Figure 3 shows plasma concentrations of lactate, K+, and Na+

measured before and after the matches. Plasma lactate and K+

were significantly increased after each match, whereas plasma

Na+ gradually decreased. Lactate concentration peaked

immediately after the match and K+ concentration 90 minutes

after the match, thereafter returning rapidly to resting levels

Figure 1 Changes in plasma myoglobin concentration andcreatine kinase activity after the rugby matches. Values are mean(SE) (n = 14). *, †Significantly different from resting state within thesame subjects (p<0.05, Wilcoxon signed ranks test).

Figure 2 Relation between the number of tackles and (A) plasmamyoglobin concentration and (B) plasma creatine kinase activity.Correlation analysis was carried out by linear regression, and thePearson correlation coefficient (r) was calculated.

Muscle damage in rugby 417

www.bjsportmed.com

on 15 December 2005 bjsm.bmjjournals.comDownloaded from

(figs 3A,B). Na+ concentration reached its lowest level 90 min-

utes after the match and thereafter returned rapidly to its

resting level (fig 3C). These results suggest that Na+ ions move

from the plasma into contracting muscle, and K+ and lactate

ions exit contracting muscle. Such an ionic exchange is typical

during intense exercise,17 indicating that the intensity of the

exercise in these matches was generally high.

DISCUSSIONThe results show that competitive rugby matches induce seri-

ous structural damage to muscle tissue, the extent of which is

highly dependent on the number of tackles. Plasma myoglobin

concentration and plasma creatine kinase activity, indirect

indicators of muscle damage, had increased significantly after

each match. Peak creatine kinase activity 24 hours after the

match was 1081 (159) U/l, which was almost as high as after

a marathon race.18

Plasma lactate and K+ concentrations had significantlyincreased after the match, whereas plasma Na+ concentrationgradually decreased. Such ionic changes are typically observedduring intense exercise,17 indicating that the intensity of exer-cise during these matches was generally high. In addition,match analysis showed that the mean durations of the workand rest periods were 21.5 (2.2) and 24.3 (3.1) seconds respec-tively. On the basis of these results, activity of high intensitywas performed intermittently more than 100 times a match.Such intermittent activity is usually accompanied by runningindependently of the mode.19 Rugby matches therefore involveconsiderable acceleration and deceleration of the centre ofmass. In particular, the hamstrings work eccentrically to slowflexion of the hip and extension of the knee during the landingphase to decelerate the centre of mass.20 21 Eccentric musclecontractions, in which the muscle lengthens as it exerts force,generate greater tension per cross sectional area of activemuscle than concentric contractions,7 resulting in consider-able structural damage to muscle tissue.7 22 Therefore muscledamage observed after rugby matches is partially caused byintermittent repetitions of intense and eccentric musclecontraction during running, especially sprinting.

Thompson et al23 reported that intermittent high intensityshuttle running for 90 minutes in an activity patternrepresentative of multiple sprint sports such as soccer, rugby,and field hockey produced muscle soreness and damage,which was mainly caused by eccentric muscle contractionsduring the landing phase in the sprinting, as reported in thisstudy. However, there is a considerable difference in the extentof muscle damage between findings in the present study andthose of Thompson et al.23 Peak plasma creatine kinase activityin our study was 1081 (159) U/l, which is about 1.4-fold higherthan that reported by Thompson et al.23 This suggests thatthere was an additional cause of the extensive muscle damagein the present study. However, peak plasma creatine kinaseactivity cannot be simply compared between the two studiesbecause of experimental differences such as the characteris-tics of the subjects.

A competitive rugby match usually involves aggressive andforceful plays with body contact such as scrums, tackles,rucks, and mauls. The tackle causes the first contact withanother player in the opposing team at every restart of anyplay in rugby matches, which is necessarily followed by othercontact plays such as rucks and mauls. Tackling is one of thebasic skills required to stop the ball carrier in the opposingteam and is common to all playing positions. In particular,tackles from in front of the tackled player are carried out tostop the ball carrier, who is travelling with great momentum,usually resulting in greater force at impact and greater kineticenergy to the tackling player. However, if the tackling playerhas the greater momentum, the tackled player receives greaterkinetic energy—that is, the tackled player receives more dam-age to his body. In any case, it is clear that the total momentumof the tackling and tackled players before body contact isredistributed between them at impact by the law of conserva-tion of energy, and also the kinetic energy produced. The play-ers involved in a tackle thus receive more or less damage totheir body with respect to each other at impact regardless ofwhether tackling or being tackled.

Therefore it is reasonable to suggest that the tackle isrepresentative of plays involving body contact in rugby, andattention should be focused on both tackling and tackledplayers. In this study, the relation between the total number oftackling and tackled plays, and the extent of muscle damageafter competitive rugby matches was investigated to help toclarify the cause of muscle damage. There was a significantcorrelation between the number of tackles and both peakmyoglobin concentration (fig 2A; r = 0.85, p<0.01) and peakcreatine kinase activity (fig 2B; r = 0.92, p<0.01). Thereforemuscle damage is dependent on the number of tackles duringa match. In addition, the small intercept values of each

Figure 3 Changes in plasma concentrations of (A) lactate, (B) K+,and (C) Na+ after the rugby matches. Values are mean (SE) (n = 14).*Significantly different from resting state within the same subjects(p<0.05, Wilcoxon signed ranks test).

418 Takarada

www.bjsportmed.com

on 15 December 2005 bjsm.bmjjournals.comDownloaded from

regression equation exhibited in fig 2 suggest that the directimpact of the tackle was the major cause of muscle damage.

The subjects were highly experienced, technically skilledrugby players (more than 10 years of playing), who werefamiliar with activity patterns found in rugby such asintermittent running or sprinting through either daily practiceor matches over many years. Contact plays such as tackles,which are also activity patterns representative of rugby, are notusually experienced at such a severe level in daily practice as incompetitive matches. Therefore, the highly technical skills andfamiliarity with typical activity patterns found in rugby, exceptfor hard tackles, in the present subjects is probably the mainreason for the reduced muscle damage induced by the repeatedeccentric muscle contractions, despite the considerableintermittent running/sprinting during the matches.

In summary, the direct impact of tackles on the body isprobably the major cause of muscle damage observed after acompetitive rugby match, in addition to the repeated eccentricmuscle contractions involved in the intermittent running/sprinting of rugby matches.

Muscle damage immediately after strenuous exercise hasbeen observed as disruptions in the A band and localised dis-solution of Z lines24 by electron microscopy,8 25 which wasfound to be associated with the structural damage of the con-tractile apparatus due to mechanical stress. This repeatedmicro-damage to muscle fibres is suggested to cause musclestrains or tears graded as injuries. In addition, the capacity ofthe damaged muscle to generate force is greatly reduced,4 14

and this reduction in muscle function not only prevents astable rugby performance during a match, but also increasesthe risk of injuries such as strained or torn ligaments of theknee joint because of increased joint laxity.26 Indeed, mostinjuries in rugby players are sprains, strains, and ruptures ofmuscles, ligaments, and joints in the lower limbs.27 Theincidence of these injuries increases with the increasednumber of tackles during a match, which clearly increases theextent of muscle damage. This is supported by the findings ofprevious studies27 28 showing that many injuries occurredwhen players were hit by each other in tackling situations.

In Japan, the competitive rugby season is between Septem-ber and January. Between seven and fourteen matches aregenerally played in league and elimination tournaments. It is ofparamount importance that players stay in good physical con-dition throughout the season for good results to be achieved.Therefore injury prevention and rapid recovery from muscledamage is very important. In general, severe muscle sorenessand decreased muscle strength are induced by unaccustomedstrenuous exercise, regardless of training status.6 However, theextent of muscle damage is significantly influenced by trainingstatus.6 29 30 Vincent and Vincent6 reported that mean peakcreatine kinase activity in trained subjects after intense weighttraining for leg muscles was about 60% lower than that inuntrained subjects, suggesting that training allows muscles tobecome more resistant to damage by exercise. In addition,damage to a trained muscle is repaired at a faster rate.6 29 30

Indeed, the second and third bouts of the same exercise oneweek after the first bout of severe unaccustomed eccentricexercise was found to induce considerably less muscle damageand loss of strength even in untrained subjects, and this train-ing effect was found to last about six weeks, suggesting a con-siderable and long lasting adaptation.4 31 32 These resultssuggest that introducing hard tackles, which are the maincause of muscle damage in rugby matches, into daily rugbytraining during the preseason may reduce the extent of muscledamage during matches in the competitive season. However,further studies are required to elucidate this possible effect.

REFERENCES1 Bruunsgaard H, Galbo H, Halkjaer-Kristensen J, et al. Exercise-induced

increase in serum interleukin-6 in humans is related to muscle damage. JPhysiol (Lond) 1997;499:833–41.

2 De Paz JA, Villa JG, Lopez P, et al. Effects of long-distance running onserum bilirubin. Med Sci Sports Exerc 1995;27:1590–4.

3 Kyrolainen H, Takala TES, Komi PV. Muscle damage induced bystretch-shortening cycle exercise. Med Sci Sports Exerc 1998;30:415–20.

4 Newhan DJ, Jones DA, Clarkson PM. Repeated high-force eccentricexercise: effects on muscle pain and damage. J Appl Physiol1987;63:1381–6.

5 Rokitzki L, Logemann E, Sagredos AN, et al. Lipid peroxidation andantioxidative vitamins under extreme endurance stress. Acta PhysiolScand 1994;151:149–58.

6 Vincent HK, Vincent KR. The effect of training status on the serumcreatine kinase response, soreness and muscle function followingresistance exercise. Int J Sports Med 1997;18:431–7.

7 Armstrong RB, Oglivie RW, Schwane JA. Eccentric exercise-inducedinjury to rat skeletal muscle. J Appl Physiol 1983;54:80–93.

8 Friden J, Sjostrom M, Ekblom B. Myofibrillar damage following intenseeccentric exercise in man. Int J Sports Med 1983;4:170–6.

9 Jones DA, Newham DJ, Round JM, et al. Experimental human muscledamage: morphological changes in relation to other indices of damage. JPhysiol (Lond) 1986;375:435–48.

10 McNeil PL, Khakee R. Disruptions of muscle fiber plasma membranes:role in exercise-induced damage. Am J Pathol 1992;140:1097–109.

11 Tiidus PM, Ianuzzo CD. Effects of intensity and duration of muscularexercise on delayed soreness and serum enzyme activities. Med SciSports Exerc 1983;15:461–5.

12 Rogers MA, Stull GA, Apple FS. Creatine kinase isoenzyme activities inmen and women following a marathon race. Med Sci Sports Exerc1985;17:679–82.

13 Nosaka K, Clarkson PM. Variability in serum creatine kinase responseafter eccentric exercise of the elbow flexors. Int J Sports Med1996;17:120–7.

14 Clarkson PM, Nosaka K, Braun B. Muscular function afterexercise-induced muscle damage and rapid adaptation. Med Sci SportsExerc 1992;24:512–20.

15 Zuliani U, Bonetti A, Franchini D, et al. Effect of boxing on some metabolicindices of muscluar contraction. Int J Sports Med 1985;6:234–6.

16 Asanuma K, Ariga T, Miyasaka N, et al. A new method forsimultaneous autoanalysis of plasma levels lactic acid and pyruvic acidby means of the oxidase system. Analysis of Biological Materials1985;8:16–24.

17 Lindinger MI, Heigenhauser GJF, McKelvie RS, et al. Blood ionregulation during repeated maximal exercise and recovery in humans.Am J Physiol 1992;262:R126–36.

18 Ostrowski K, Schjerling P, Pedersen BK. Physical activity and plasmainterleukin-6 in humans: effect of intensity of exercise. Eur J Appl Physiol2000;83:512–15.

19 Nicholas CW. Anthropometric and physiological characteristics of rugbyunion football players. Sports Med 1997;23:375–96.

20 Friden J, Serger J, Ekblom B. Sublethal muscle fibre injuries afterhigh-tension anaerobic exercise. Eur J Appl Physiol 1988;57:360–8.

21 Williams K. Biomechanics of running. In: Terjung R, ed. Exercise andsports science reviews. London: Macmillan, 1985;13:389–441.

22 Saxton JM, Donnelly AE, Roper HP. Indices of free-radical-mediateddamage following maximum voluntary eccentric and concentric muscularwork. Eur J Appl Physiol 1994;68:189–93.

23 Thompson D, Nicholas CW, Williams C. Muscular soreness followingprolonged intermittent high-intensity shuttle running. J Sports Sci1999;17:387–95.

24 Oglivie RW, Armstrong RB, Baird KE, et al. Lesions in the rat soleus musclefollowing eccentrically based exercise. Am J Anat 1988;182:335–46.

25 Newhan DJ, McPhail G, Mills KR, et al. Ultrastructural changes afterconcentric and eccentric contractions of human muscle. J Neurol Sci1983;61:109–22.

26 Nicholas JA. Injuries to knee ligaments. Relationship to looseness andtightness in football players. JAMA 1970;212:2236–9.

27 Garraway WM, Lee AJ, Hutton SJ, et al. Impact of professionalism oninjuries in rugby union. Br J Sports Med 2000;34:348–51.

28 Bird YN, Waller AE, Marshall SW, et al. The New Zealand rugby injuryand performance project. V. Epidemiology of a season of rugby injury.Br J Sports Med 1998;32:319–25.

29 Clarkson PM, Tremblay I. Exercise-induced muscle damage, repair andadaptation in humans. J Appl Physiol 1988;65:1–6.

30 Paul GL, DeLany JP, Snook JT, et al. Serum and urinary markers ofskeletal muscle tissue damage after weight lifting exercise. Eur J ApplPhysiol 1989;58:786–90.

31 Ebbeling CB, Clarkson PM. Muscle adaptation prior to recoveryfollowing eccentric exercise. Eur J Appl Physiol 1990;60:26–31.

32 Jones DA, Newman DJ. The effect of training on human muscle painand damage [abstract]. J Physiol (Lond) 1985;365:76P.

Take home message

Competitive rugby matches cause serious structuraldamage to muscle tissue, the extent of which depends onthe number of tackles. The direct impact of the tackle on thebody is the major cause of muscle damage.

Muscle damage in rugby 419

www.bjsportmed.com

on 15 December 2005 bjsm.bmjjournals.comDownloaded from

③参考資料３

Force overestimation during tourniquet-induced transient occlusion of the

brachial artery and possible underlying neural mechanisms

Yudai Takarada a, Daichi Nozaki b, Masato Taira c,d,*aFaculty of Sport Sciences, Waseda University, Saitama 359-1192, Japan

bDepartment of Rehabilitation for the Movement Functions, Research Institute of National Rehabilitation Center for

Persons with Disabilities, Saitama 359-8555, JapancNihon University, Advanced Research Institute for the Sciences and Humanities, Tokyo 102-8251, Japan

dDivision of Applied System Neuroscience, Advanced Medical Research Center,

Nihon University Graduate School of Medical Science, 30-1 Ohyaguchi-Kamimachi Itabashi, Tokyo 173-8610, Japan

Received 13 August 2005; accepted 11 October 2005

Available online 14 November 2005

Abstract

Avascular occlusion by a tourniquet inflated at the proximal end of the upper arm is suggested to affect the estimation of exertion force level. In

the first part of this study, subjects were asked to estimate the isometric force exerted by the occluded hand with that of the other hand (matching

experiment).We found that the perceived force with arterial occlusion was always overestimated. To examine the underlying neural mechanism for

this phenomenon, in the second part, the somatosensory evoked potentials (SEPs) and nerve action potential (NAP) were recorded following

electrical median nerve stimulation with or without arterial occlusion. Moreover, the maximum motor response (M response) to median nerve

stimuli at the axilla was recorded from the skin surface of the thenar eminence muscle of the hand during with arterial occlusion. The N20 of SEP

and NAP at Erb’s point were unaffected by the arterial occlusion, and theM responsewas also unchanged. These results suggest that the tourniquet-

induced transient occlusion of the brachial artery does not seriously affect median nerve function. Thus, it is likely that the primary responsible

factor for the overestimation of perceived force exertion during arterial occlusion is the centrally generated motor command as previously

hypothesized byMcCloskey [McCloskey, D.I., Ebeling, P., Goodwin, G.M., 1974. Estimation of weights and tensions and apparent involvement of

a ‘‘sense of effort’’. Exp Neurol. 42, 220–232; McCloskey, D.I., 1978. Kinesthetic sensibility. Physiol. Rev. 58, 763–820; McCloskey, D.I., 1981.

Corollary discharge and motor commands and perception. In: Brookhart, J.M., Mountcastle, V.B. (Eds.), Handbook of Physiology. American

Physiological Society, Bethesda, pp. 1415–1447].

# 2005 Elsevier Ireland Ltd and the Japan Neuroscience Society. All rights reserved.

Keywords: Vascular occlusion; Force perception; SEP; M response; Central motor command; Median nerve

www.elsevier.com/locate/neures

Neuroscience Research 54 (2006) 38–42

1. Introduction

It is a common experience to feel a weight heavier than

normal when we carry it long enough and when the muscles

supporting the weight have been fatigued. McCloskey et al.

(1974) have actually shown that weight is determined to be

heavier than the usual under experimental muscle fatigue

condition by a force matching task. In their study, they also

showed that weight is overestimated during the inhibition of

the motoneurons of the agonist muscle by vibration of its

antagonist muscle. Thus, they suggested that the centrally

* Corresponding author. Tel.: +81 3 3972 8111x2231; fax: +81 3 3972 8292.

E-mail address: masato@med.nihon-u.ac.jp (M. Taira).

0168-0102/$ – see front matter # 2005 Elsevier Ireland Ltd and the Japan Neuro

doi:10.1016/j.neures.2005.10.004

generated motor command (i.e., effort) is involved in the

sensation of heaviness rather than factors in the peripheral

nerves and/or muscles (see reviews in McCloskey, 1978,

1981).

In preliminary experiments on handgrip contractions with

the proximal end of an upper arm occluded by a tourniquet at

250 mmHg, we noticed that subjects reported the need for

greater force to keep the same force level during the tourniquet

was inflated. However, thesewere just objective reports. In this

study, we first examined how arterial occlusion affects the

perception of handgrip force by the force matching experi-

ment. The results showed that the subjects always over-

estimated their exertion force during arterial occlusion.

McCloskey’s idea (1974, 1978, 1981) is thought to be the

underlying neural mechanisms for this, however the simplest

science Society. All rights reserved.

Y. Takarada et al. / Neuroscience Research 54 (2006) 38–42 39

explanation is that ischemia of the forearm and/or the

tourniquet itself cause mechanical deformation of the nerves

and interfere with conduction of the peripheral nerve. Previous

studies showed that the tourniquet-induced ischemia

diminishes amplitudes of the early-latency somatosensory

evoked potentials (SEPs) (Yamada et al., 1981; Nishihira et al.,

1996). Thus, in the second part of the present study, we

investigated the early cortical SEPs to median nerve stimuli at

the wrist (SEPs experiment), and M waves to median nerve

stimuli at the axilla (M response experiment) during arterial

occlusion by a tourniquet inflated around the upper arm, in

order to elucidate the possible neural mechanisms underlying

such a psychological effect of arterial occlusion on the

perceived force exertion (matching experiment).

2. Methods

2.1. Matching experiment

2.1.1. Subjects

Eight right-handed volunteers (seven men and one woman) with no history

of neurological or other disease participated in the experiments. Their mean age

was 19.8 � 0.46 years old (mean � S.D., range 19–22). All of the subjects were

informed well about the experimental procedures to be performed as well as the

purpose of the study, and their written informed consents were obtained. The

study was conducted in accordance with the Declaration of Helsinki and

approved by the Ethical Committee of Faculty of Sport Sciences, Waseda

University.

2.1.2. Procedure

The subjects were placed in a sitting position with their upper body upright

and their upper arm inclined at about 458 in front of the body with the aid of anarmrest. To measure handgrip force, handgrip devices with a strain gage (KFG-

5-120-C1-16; Kyowa Electronic Instruments Co. Ltd., Tokyo, Japan) were held

in the subject’s right and left hands. The measured force was amplified (AD240-

A; TEAC Instruments Co., Kawasaki, Japan) and inputted to a visual feedback

system (IBM T40) to show the subjects the exerted force and the predetermined

target force level on the display. The force was digitized (200 samples/s) and

recorded on the hard disk of a data acquisition computer (Macintosh Power-

Book G4: M8858J/A).

Before matching measurement (see below), the maximal voluntary con-

traction (MVC) was measured to calculate 20%, 40%, 60% and 80% MVC,

each subject was asked to squeeze the handgrip device with his maximum effort

three times with a 120–180 s rest between squeezing. The values of the force

exerted were averaged over the middle 1 s during muscle contraction.

During the matching measurements, only the force exerted in the reference

(right) hand was displayed on the monitor. The subjects were asked to align the

force exerted by the reference hand with a predetermined target force by visual

feedback, and simultaneously to contract the indicator (left) hand so as to

‘‘make both hands the same’’, i.e., without visual feedback. In one measurement

session, this bilateral isometric contraction (approximately 3–5 s) was per-

formed three times with a 10–15 s rest in between contractions at a given target

force (20%, 40%, 60% or 80% MVC) with or without arterial occlusion. There

was a rest for 120–180 s between each session and the order of sessions was

randomized for each subject. The arterial occlusion was produced by a

tourniquet, which was attached at the proximal end of the right upper arm.

A pressure of 250 mmHg started to be applied by pneumatic inflation approxi-

mately 15 s before each muscle contraction measurement session, which was

maintained throughout the each measurement session, and was released imme-

diately after the end of the measurement session. Thus, the arterial occlusion in

one measurement session was lasted for 29–45 s.

2.1.3. Analysis

To normalize the differences in exerted handgrip force between the refer-

ence hand and the indicator hand, matching value [MV (%)] was calculated.

MV ð%Þ ¼

Handgrip force of the indicator hand

� handgrip force of the reference hand

Handgrip force of the reference hand� 100

2.2. SEPs experiment and M response experiment

2.2.1. Subjects

Nine right-handed volunteers (seven men and two women) with no history

of neurological or other disease participated in the SEPs experiments. Their

mean age was 20.6 � 0.74 years old (mean � S.D., range 19–22).

Six (five men and one woman) of the nine subjects participated in the M

response experiments. Their mean agewas 20.0 � 0.63 years old (mean � S.D.,

range 19–22). All of the subjects were informed well about the experimental

procedure to be performed as well as the purpose of the study, and their written

informed consents were obtained. The study was conducted in accordance with

the Declaration of Helsinki and approved by the Ethical Committee of Faculty

of Sport Sciences, Waseda University.

2.2.2. Procedure: SEPs experiment

Electrical stimuli (duration, 0.2 ms; 5 Hz at random; intensity, 110–115% of

muscle twitch threshold) were delivered to the right median nerve at the wrist

using flat-surfaced disk electrodes by a stimulator (DiaMedical, DPS-1300) and

an isolator (DPS-107). The SEP signals and nerve action potentials (NAP) of the

median nerve were recorded by silver discs (impedance: <5 V) attached to the

skin with an EEG paste. The exploring-referential electrodes were placed on C3

(2 cm posterior to Cz and 7 cm lateral to midline in the hemisphere contralateral

to the stimulated side)–Ai (earlobe in the hemisphere contralateral to the

stimulated side), and Erb’s point (contralateral to the stimulated side)–Erb’s

point (ipsilateral to the stimulated side) (Mauguiere et al., 1999). Signals

(analysis time: 40 ms) were amplified (NEC BIOTOP 6R12) using a bandpass

filter of 20 Hz–3 kHz, digitized (National Instruments, NI DAQCard-6036E,

2 kHz sampling) and averaged 600 times by a personal computer (Toshiba

Dynabook V9/W14DEW).

In one recording session, the SEPs and NAP to the median nerve stimuli

were recorded with or without vascular occlusion at a pressure of 250 mmHg

applied by a tourniquet at the proximal end of the upper arm. The duration of

each recording condition was 120 s (thus 600 sweeps were collected at each

condition) followed by a 60 s rest. The session was repeated two times with

3 min rest interval.

2.2.3. Analysis: SEPs experiment

SEP components were labeled according to a polarity–latency convention

(negative waves, N; positive waves, P) followed by a number indicating the

measured peak latency in milliseconds. In analyzing the amplitude, the peak-to-

peak amplitude was measured from the peak of the preceding reversed-polarity

component. The amplitude of NAP recorded from the Erb’s point (Erb–Erb) was

measured from its onset to the major negative peak.

2.2.4. Procedure: M response experiment

Electrical stimuli (duration, 0.2 ms; 3 Hz) were delivered by the stimulator

and the isolator to the right median nerve at the axilla using flat-surfaced disk

electrodes (12 mm wide and 46 mm long) with the cathode 20 mm proximal to

the anode.

The motor response in the EMG was recorded from the skin surface of the

thenar eminence muscle in the right hand. Surface electrodes (diameter 8 mm)

were placed over the bellies of these muscles, with a constant interelectrode

distance of 20 mm. A reference electrode was fixed on the skin overlying the

lateral epicondyle near the elbow joint of the right arm. The maximum motor

response (M response) was elicited by the supramaximal stimulation of the

median nerve at the axilla.

The EMG signals were amplified, filtered through both low (20 Hz) and

high (1 kHz) cut filters, digitized (National Instruments, NI DAQCard-6036E,

2 kHz sampling), and stored in the personal computer (Toshiba Dynabook V9/

Y. Takarada et al. / Neuroscience Research 54 (2006) 38–4240

W14DEW). M response was recorded with and without arterial occlusion. The

manner of recording session was the same as that of the SEP experiment.

2.2.5. Analysis: M response experiment

TheMwaves in response to 10 stimuli of the median nervewere averaged in

each subject, and the latencies of these averaged waves were measured from the

stimulus to each peak. Also, the time integral (mV ms) under the full-wave

rectified M wave recorded by the surface electrodes was adopted as an index of

the M response.

3. Results

3.1. Matching experiment

We asked the subjects to determine the magnitude of the

handgrip force exerted by the reference (right) hand by

producing a brief matching contraction on the indicator (left)

hand, to estimate numerically the subjective effort required to

exert the handgrip force by the reference hand.

When the handgrip force exerted by the indicator hand was

plotted against that by the reference hand, a high correlation

was observed (r = 0.99l; Fig. 1). Thus, the handgrip force

exerted by the indicator can be an indicator of that by the

reference hand during bilateral isometric handgrip contrac-

tions. Such a linear relation was also held during the handgrip

force exertion by the reference hand with arterial occlusion

(r = 0.98). Although there was no difference in the handgrip

force of reference hand between with and without arterial

occlusion conditions (n.s., Student’s paired t-test), statistically

significant increases in that of the indicator hand were

observed in the occluded condition at all four different levels of

the target force ( p < 0.05, Student’s paired t-test). In order to

elucidate whether the transient arterial occlusion change

precision in perceiving force exertion, we compared the

standard deviation of handgrip force of the indicator and

reference hand between two conditions (with and without

arterial occlusion). In both handgrip forces, there was no

Fig. 1. Relationship between reference and indicator handgrip forces with

(filled circle) or without arterial occlusion (unfilled circle). The values of the

reference and indicator handgrip forces were normalized to those of the

maximal voluntary contractions (MVCs) on the reference and the indicator

hands, respectively, and means with S.E. (n = 8) were plotted.

statistically significant difference in the standard deviation

between two conditions (n.s., Student’s paired t-test). Thus, the

transient arterial occlusion did not affect on precision in

perceiving force exertion.

Fig. 2 shows the MVs in the control and occluded conditions

at each level of the target force, which were calculated by the

difference of exerted handgrip force between by the reference

hand and by the indicator hand (see Section 2). The average

MVs in the control condition at the four different target levels

were 12.7 � 7.3%, 9.7 � 7.8%, 6.5 � 5.4% and 1.1 � 5.2%

(mean � S.E.), respectively, and those in the occluded

condition were 45.0 � 10.1%, 30.9 � 8.9%, 16.8 � 5.2%

and 6.9 � 4.6% (mean � S.E.), respectively. These results

show that the handgrip forces exerted by the indicator hand

increased significantly at all levels of the target force during

handgrip contractions of the reference hand when combined

with arterial occlusion ( p < 0.05, Student’s paired t-test),

suggesting that an arterial occlusion led to the overestimation of

the exerted force.

3.2. SEPs experiment

We analyzed the amplitudes and latencies of NAP recorded

from the Erb’s point (Erb–Erb), and SEPs (N20 and P23

component) recorded from the parietal area (C3–Ai). Fig. 3

shows the SEPs and NAP waveforms, which were averaged for

all the subjects (n = 9) to the median nerve stimuli in the control

and occluded condition. The peak latencies of NAP, N20 and

P23 in the control condition were 8.6 � 0.3 ms, 19.1 � 0.2 ms

and 22.8 � 0.5 ms, respectively. The peak latencies of NAP,

N20 and P23 in the occluded condition were 8.7 � 0.3 ms,

19.1 � 0.2 ms and 22.7 � 0.6 ms, respectively. These results

showed that the peak latencies of NAP, N20 and P23 of SEP

components were unchanged in the occluded condition. The

amplitude of the N20 component recorded from the parietal

area was unchanged; that of the P23 components recorded from

Fig. 2. Matching values with (hatched square) or without (square) arterial

occlusion at four different target forces (20%, 40%, 60% and 80% MVC).

Values are expressed as means with S.E. (n = 8). The asterisk represents

statistically significant differences between the control and occluded conditions

( p < 0.05, Student’s paired t-test).

Y. Takarada et al. / Neuroscience Research 54 (2006) 38–42 41

Fig. 3. Nerve action potential at the Erb’s point (NAP; A) and short-latency

SEPs (N20 and P23; B) at C3 during median nerve stimuli with (filled circle) or

without arterial occlusion (unfilled circle) in one subject.

the parietal area was significantly decreased in the occluded

condition ( p < 0.01, Student’s paired t-test). The amplitude of

the P23 component decreased from 6.49 � 0.59 mV in the

control condition to 4.01 � 0.65 mV in the occluded condition.

Fig. 4. M waves following median nerve simulation during rest (dashed line)

and arterial occlusion (solid line) in one subject.

These results showed that the transient arterial occlusion does

not significantly affect the cortical SEP component, N20 and

median nerve function.

3.3. M response experiment

The amplitudes of M waves recorded from the thenar

eminence muscle in the control and occluded condition were

7.01 � 3.01 mVand 6.67 � 2.70 mV, respectively (Fig. 4). The

time integral under the full-wave rectifiedMwave in the control

and occluded conditions were 6.34 � 2.51 mV ms and

5.68 � 1.87 mV ms, respectively. No statistically significant

differences between the control and occluded conditions were

observed in the amplitudes and the time integral of M waves

(n.s., Student’s paired t-test), showing that such an arterial

occlusion around the upper arm did not produce any substantial

change in the M response.

4. Discussion

It is a common experience to feel a weight heavier than

normal when we carry it long enough and when the muscles

supporting the weight have been fatigued. In preliminary

experiments, arterial occlusion at the proximal end of the upper

arm caused the same effect on handgrip contractions; we

noticed that subjects reported the need for greater force to keep

the same force level during the arterial occlusion. The force

matching experiment in this study revealed that arterial

occlusion applied around the upper arm by a tourniquet

elevates the perceived magnitude of the force exerted during

handgrip contractions (Figs. 1 and 2). Although there was no

difference in the exerted force by reference hand between with

and without arterial occlusion conditions, the exerted force by

indicator hand revealed that subjects felt exerting more force

during arterial occlusion. McCloskey et al. (1974) hypothesized

that the centrally generated motor command (i.e., effort) is

involved in the sensation of heaviness. Thus, this mechanism

may similarly operate in the overestimation of exerted force

during muscular contractions with arterial occlusion in this

study. At the same time, however, we cannot deny the

possibility that ischemia of the forearm and/or the tourniquet

itself cause the mechanical deformation of the nerves and

interferes with conduction of the peripheral nerve.

Indeed, several previous studies demonstrated that a

tourniquet-induced vascular occlusion of the upper arm

markedly attenuated the amplitudes of early-latency SEPs

and Erb’s potential to median nerve stimuli at the wrist

(Yamada et al., 1981; Nishihira et al., 1996). In this study,

however, there were no significant changes in the peak latencies

and amplitudes of NAP (Fig. 3A), and of the early-latency SEP,

N20 (Fig. 3B) to the median nerve stimuli during arterial

occlusion. This inconsistency of the effect of vascular occlusion

on median nerve function may be mainly caused by the

difference in the duration of occlusion, rather than the degree of

occlusion. In previous studies, the duration of vascular

occlusion was relatively long [from 24 min (Yamada et al.,

1981) to 30 min (Nishihira et al., 1996)], however, duration was

Y. Takarada et al. / Neuroscience Research 54 (2006) 38–4242

not exceeded more than 150 s in our experiments. In other

words, vascular occlusion with a short duration of not more

than 150 s does not induce any deterioration of median nerve

function, whatever inflation pressure magnitude is applied by a

tourniquet. This conclusion is also reinforced by the results on

M response experiment in which no significant changes were

observed in the peak-to-peak amplitudes and the time integral

of M waves recorded from the thenar eminence muscle to

median nerve stimuli at the axilla during arterial occlusion

(Fig. 4).

When the tourniquet was inflated to 250 mmHg, subjects felt

pressure on the skin and a light tingling sensation in deep tissue,

however, these sensations were immediately disappeared with

deflation. It is well known that cutaneous afferent nerves

innervate to spinal motoneurons through excitatory or

inhibitory interneurons. Previous studies (Goodwin et al.,

1972; McCloskey et al., 1974; also see reviews in McCloskey,

1978, 1981) shown that, if the spinal motoneurons of the

agonist muscle were inhibited by the spindle afferents of the

vibrated antagonist muscle, subjects overestimated the per-

ceived force exertion in the matching test. Thus, it is suggested

that the same neural mechanisms are involved in the present

study; inhibition of the spinal motoneurons by the somatic

sensations through the cutaneous afferent nerves might be one

of the underlying neural mechanisms for the overestimation of

force. It is, also, fact that the amplitude of the P23 component

significantly decreased during arterial occlusion ( p < 0.01,

Student’s paired t-test). In general, subsequent components to

N20 at 20–30 ms, including P23, are thought to reflect activities

of higher cortical areas, however, the generator locations of

those components have not been confirmed yet (Allison et al.,

1991; Kakigi, 1994; Waberski et al., 1999; Hoshiyama and

Kakigi, 2001). Thus, the attenuation of P23 might be related to

the neural aspects of the overestimation of perceived force

exertion, however, we need further experiments to clarify the

underlying neural mechanisms.

The results presented above show that a tourniquet-induced

transient occlusion of the brachial artery does not inhibit the

normal efferent and afferent functions of the median nerve.

Thus, it is likely that the primary responsible factor for the

overestimation of perceived force exertion during arterial

occlusion is the centrally generated motor command as

previously hypothesized by McCloskey (1978, 1981).

The present study does not indicate directly the source of the

centrally generated motor command, however, it is reasonable

to assume that the activity of MI is closely involved in such an

increase in the perceived magnitude of handgrip force during

arterial occlusion. To investigate this possibility, further studies

are being carried out with functional magnetic resonance

imaging (fMRI) and trans cranial magnetic stimulation (TMS).

Acknowledgments

This study was supported by Academic Frontier Project for

Private Universities: ‘‘Brain Mechanisms for Cognition,

Memory and Behavior’’ at Nihon University matching fund

subsidy fromMEXTand a Grant-in-Aid for Scientific Research

on Priority Areas-Advanced Brain Science project-fromMEXT

(No. 16500425).

References

Allison, T., McCarthy, G., Wood, C.C., Jones, S.J., 1991. Potentials evoked in

human and monkey cerebral cortex by stimulation of the median nerve. A

review of scalp and intracranial recordings. Brain 114, 2465–2503.

Goodwin, G.M., McCloskey, D.I., Mitchell, J.H., 1972. Cardiovascular and

respiratory responses to changes in central command durin isometric

exercise at constant muscle tension. J. Physiol. Lond. 226, 173–190.

Hoshiyama, M., Kakigi, R., 2001. Correspondence between short-latency

somatosensory evoked brain potentials and cortical magnetic fields follow-

ing median nerve stimulation. Brain Res. 908, 140–148.

Kakigi, R., 1994. Somatosensory evoked magnetic fields following median

nerve stimulation. Neurosci. Res. 20, 165–174.

McCloskey, D.I., 1978. Kinesthetic sensibility. Physiol. Rev. 58, 763–820.

McCloskey, D.I., 1981. Corollary discharge and motor commands and percep-

tion. In: Brookhart, J.M., Mountcastle, V.B. (Eds.), Handbook of Physiol-

ogy. American Physiological Society, Bethesda, pp. 1415–1447.

McCloskey, D.I., Ebeling, P., Goodwin, G.M., 1974. Estimation of weights and

tensions and apparent involvement of a ‘‘sense of effort’’. Exp. Neurol. 42,

220–232.

Mauguiere, F., Allison, T., Babiloni, C., Buchner, H., Eisen, A.A., Goodin, D.S.,

Jones, S.J., Kakigi, R., Matsuoka, S., Nuwer, M., Rossini, P.M., Shibasaki,

H., 1999. Somatosensory evoked potentials. The International Federation of

Clinical Neurophysiology. Electroencephalogr. Clin. Neurophysiol. Suppl.

52, 79–90.

Nishihira, Y., Araki, H., Funase, K., Imanaka, K., Suzuki, J., Takemia, T., 1996.

Somatosensory evoked potentials following voluntary movement during

upper arm compression. Electromyogr. Clin. Neurophysiol. 36, 21–28.

Waberski, T.D., Buchner, H., Perkuhn, M., Gobbele, R., Wagner, M., Kucker,

W., Silny, J., 1999. N30 and the effect of explorative finger movements: a

model of the contribution of the motor cortex to early somatosensory

potentials. Clin. Neurophysiol. 110, 1589–1600.

Yamada, T., Muroga, T., Kimura, J., 1981. Tourniquet-induced ischemia and

somatosensory evoked potentials. Neurology 31, 1524–1529.