腸結核における遺伝子診断の有用性と問題点―病理組織学的所見および結核菌培養との比較

Evaluation of the Polymerase Chain Reaction and the Restriction Fragment Length Polymorphism Method for the Diagnosis of Intestinal Tuberculosis

要旨　腸結核における遺伝子診断の有用性について，17例の腸結核患者（非手術例12例，手術例5例）を対象に検討した．方法は生検および外科手術標本のパラフィンブロックから薄切標本を作製してDNAを抽出し，polymerase chain reaction（PCR）で増幅した後，制限酵素で消化してrestriction fragment length polymorphism（RFLP）のパターンから結核菌を同定した．生検組織標本においては12例中8例が陽性を示し，感度は66.7％であった．それに対して，非腸結核症例29例ではいずれも陰性で，特異度は100％，有用度は90.2％であった．従来の生検組織診断と結核菌培養との組み合わせにより，12例中11例（91.7％）で結核の診断が可能であり，PCR-RFLP法は腸結核の補助診断法として有用と考えられた．一方，外科手術標本では5例いずれも陰性で，長時間ホルマリン液に浸したことが原因と考えられた．本法の感度を更に向上させるためには新鮮組織標本での検討が必要と思われた．

　To evaluate the value of molecular biological techniques for the diagnosis of mycobacterial infection, 17 cases of intestinal tuberculosis (12 non-operated cases and five operated ones) were studied. Deoxyribonucleic acid (DNA) was extracted from paraffin blocks of biopsy and surgically resected materials, polymerase chain reaction (PCR) was performed. In addition, PCR products were resolved by the restriction enzyme, and Mycobacterium tuberculosis was detected by the pattern of restriction fragment length polymorphism (RFLP) using electrophoresis. In the biopsy specimens, eight of 12 specimens of non-operated intestinal tuberculosis were positive, and the sensitivity of PCR-RFLP method was 66.7%. On the other hand, 29 biopsy specimens of non-intestinal tuberculosis were all negative, so the specificity was 100% and the accuracy was 90.2%. Combined PCR-RFLP method and the conventional methods (histology and culture) established the diagnosis in 11 of 12 cases (91.7%). Consequently, the PCR-RFLP method is thought to be a useful procedure for the diagnosis of intestinal tuberculosis. In the surgically resected specimens, however, five specimens were all negative, and the sensitivity was 0%. The reason for the discrepancy of sensitivity between biopsy specimens and surgically resected ones was thought to lie in the fact that surgical materials had been preserved by formalin for longer periods (3～7 days) compared with biopsy ones (1 day). Because DNA is injured by formlin, it is necessary to examine fresh and non-fixed materials in order to obtain an improvement of the sensitivity of this procedure.