NSAIDs的不良反应 北医三院药剂科 段京莉

引言 人类使用非甾体抗炎药（ nonsteroidal anti-inflammatory drugs , NSAIDs）已有100多年的历史； 全球每天约有3千万人使用NSAIDs，仅美国每年就有7～10亿张NSAIDs处方。在国内，NSAIDs销量仅次于抗感染药，位居第二； NSAIDs致不良反应的发生率之高，同样不容忽视。在所有有关药物不良反应的报道中，NSAIDs占25％。 预防NSAIDs 的不良反应及如何改善抗炎药物治疗已成为医药工作者共同关注的课题！

NSAIDs百年历程 1960’- 1998年 成功克隆环氧酶COX-1的同工酶COX-2 第一个NSAIDs 阿司匹林诞生 全球第一个选择性COX-2抑制剂- 塞来昔布，获FDA批准在美国上市 1.向继洲主编. 药理学.科学出版社. 2.Data on file. Pfizer Inc, New York, NY. •塞来昔布在中国获得

NSAIDs分类 COX-2抑制剂 传统NSAIDs 昔布类 塞来昔布 （西乐葆） 苯胺类 水杨酸类 阿司匹林 有机酸类 COX-2抑制剂于 （泰诺林／百服宁) 非那西林 奈基烷酸 奈丁美酮 （瑞力芬） 水杨酸类 阿司匹林 有机酸类 COX-2抑制剂于 99年后 应用于临床 吲哚类（酪酸） 消炎痛（吲哚美辛） 舒林酸（奇诺力） 昔康类 美洛昔康 （莫比可） 丙酸.苯乙酸 奈普生 布洛芬（芬必得） 双氯芬酸（扶他林） COX-2抑制剂：选择性COX-2抑制剂 传统NSAIDs：非选择性COX-2抑制剂

近年撤市的NSAIDs 1982年上市仅4个月的苯恶洛芬(Benoxaprofen)、 1983年上市29个月的佐灭酸(Zomepirac)、 1987年上市16个月的舒洛芬(Suprofen)、 1998年上市11个月的溴芬酸(Bromfenac) 2004年9月30日因患者服用环氧化酶COX-2抑制剂万络导致心 血管疾病，默沙东公司宣布在全球范围内撤回 万络撤出市场的罗非昔布(Rofecoxib)。 其他药物发展史上是极为罕见的。

2005年4月美国FDA最新声明 目前市场上所有的非甾体抗炎药都可能存在心血管问题，而不是一两个药独有的问题 要求药品生产厂家对OTC药物说明书进行修改，必须增加“有潜在的皮肤反应”的警告。对处方药说明书中加入可能引发心血管事件和胃肠道出血事件的黑框警示 列举了可能产生心血管风险的21种非甾体类抗炎镇痛药，这些药中就包括我们比较常用的双氯芬酸钠、布洛芬、塞来昔布等, http://www.FDA.gov/bbs/topics/news/2005/NEW01171.html

NSAIDs的不良反应 胃肠道不良反应 血小板功能和出血 心血管不良反应 肾功能损伤 肝功能损害 变态反应 其它 The safety profiles of the COX-2–specific inhibitors have been compared with those of traditional NSAIDs and placebo.

NSAIDs引起的胃部病变 1. 胃部不适 2. 有症状的胃溃疡 3. 重度GI副反应(PUB) － 穿孔(perforation) － 溃疡(ulcers) － 出血(bleeding)

NSAIDs造成的溃疡发生率约20% OA或RA患者服用传统NSAIDS≥6个月，或因胃肠道不良反应停用 Cheatum, et al. Clin Ther.1999;21:992–1003.

COX-2 假说 (1990s) + Normal Tissue Inflammation Site Arachidonic Acid Cytokines Growth factors + COX-1 Constitutive COX-2 Inducible COX-2 Inhibitors NSAIDs Physiolgical Prostaglandin Production Pathological Prostaglandin Production Normal Functions Inflammation, pain, fever

GI Outcomes With Coxibs: Study Designs VIGOR* (N=8076) CLASS† (N=7982) Celecoxib 400 mg bid (2x max chronic dose) Rofecoxib 50 mg qd Drug Yes (21%) No ASA (≤325 mg/d) Ibuprofen 800 mg tid Diclofenac 75 mg bid Naproxen 500 mg bid Comparator OA (72%), RA (28%) RA Patients Duration Median 9 months Symptomatic + complicated ulcers Complicated upper GI events 2° end point Complicated ulcers Clinical upper GI events 1° end point GI safety of the 2 currently available coxibs, rofecoxib and celecoxib, was evaluated in comparison with NSAIDs in 2 separate randomized clinical trials, VIGOR and CLASS, respectively. There are 4 major differences between the designs of these trials: Characteristics of patient populations Comparator NSAIDs Use of concomitant medication Choice of end point Whereas all patients in VIGOR had RA, 72% of patients in CLASS had OA and only 28% had RA. The GI tolerability of rofecoxib was compared with that of naproxen in VIGOR, while the CLASS trial used 2 other comparator NSAIDs, diclofenac and ibuprofen. Patients with cardiovascular (CV) risk factors enrolled in CLASS were allowed to take aspirin (≤325 mg/d), while patients who required or who had been receiving aspirin were excluded from the VIGOR trial. In VIGOR, the primary end point was confirmed clinical UGI events, including gastroduodenal perforation or obstruction, UGI bleeding, and symptomatic gastroduodenal ulcers. Complicated UGI events (perforation, obstruction, and severe UGI bleeding) were included in the secondary end point of the VIGOR trial. UGI ulcers and ulcer complications (perforation, gastric outlet obstruction, and bleeding) were the primary end point in CLASS. Symptomatic combined with complicated ulcers were chosen as the secondary end point in this trial. Bombardier C, Laine L, Reicin A, et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. N Engl J Med. 2000;343:1520-1528. Food and Drug Administration. Celebrex: Arthritis Advisory Committee Meeting, Gaithersburg, Md; February 7, 2001. http://www.fda.gov/ohrms/dockets/ac/01/slides/3677s1.htm. Food and Drug Administration. Vioxx: Arthritis Advisory Committee Meeting, Gaithersburg, Md; February 8, 2001. http://www.fda.gov/ohrms/dockets/ac/01/slides/3677s2.htm. Silverstein FE, Faich G, Goldstein JL, et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: the CLASS study: a randomized controlled trial. Celecoxib Long-term Arthritis Safety Study. JAMA. 2000;284:1247-1255. * Vioxx Gastrointestinal Outcomes Research. † Celecoxib Long-Term Arthritis Safety Study. Bombardier et al. N Engl J Med. 2000;343:1520; Silverstein et al. JAMA. 2000;284:1247; FDA Advisory Committee Meeting, 2001, Gaithersburg, Md; FDA Arthritis Advisory Committee, 2001, Gaithersburg, Md.

GI Outcomes of VIGOR Study 1° End Point—Clinical UGI Event 2° End Point—Complicated UGI Event 1.5 5.0 Naproxen Naproxen 4.0 1.0 3.0 Cumulative incidence (%) Cumulative incidence (%) 2.0 0.5 Rofecoxib Rofecoxib The VIGOR trial was designed to evaluate the GI safety of rofecoxib (50 mg qd) compared with the NSAID naproxen (500 mg bid). The cumulative incidence of perforation, symptomatic ulcers, and GI bleeding (PUBs) was 2.08 events per 100 patient-years with rofecoxib and 4.49 with naproxen. The relative risk of sustaining a confirmed clinical UGI event with rofecoxib was significantly decreased compared with naproxen (RR=0.46; P<0.001). This difference corresponds to a 54% relative risk reduction (RRR). Regarding the secondary end point of VIGOR, the cumulative incidence of complicated PUBs was 0.59 events per 100 patient-years with rofecoxib and 1.37 with naproxen, a difference that accounts for the 57% RRR with rofecoxib (RR=0.43; P=0.005). In conclusion, in patients with RA, treatment with rofecoxib is associated with significantly fewer clinically important UGI events than treatment with naproxen, a nonselective NSAID. 1.0 (RR=0.46; P<0.001) (RR=0.43; P=0.005) 0.0 2 4 6 8 10 12 2 4 6 8 10 12 Months of follow-up Months of follow-up Bombardier et al. N Engl J Med. 2000;343:1520; FDA Advisory Committee Meeting, 2001. Gaithersburg, Md. Bombardier C, Laine L, Reicin A, et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. N Engl J Med. 2000;343:1520-1528. Food and Drug Administration. Vioxx: Arthritis Advisory Committee Meeting, Gaithersburg, Md; February 8, 2001. http://www.fda.gov/ohrms/dockets/ac/01/slides/3677s2.htm.

2° End Point—Symptomatic Ulcers/ Ulcer Complications GI Outcomes of CLASS 1° End Point—Ulcer Complications 2° End Point—Symptomatic Ulcers/ Ulcer Complications Celecoxib 400 mg bid Diclofenac 75 mg bid Ibuprofen 800 mg tid 4 2 3 % of patients % of patients 1 2 CLASS was designed to compare the UGI safety of celecoxib (400 mg bid) with 2 NSAIDs: diclofenac (75 mg bid) and ibuprofen (800 mg tid). Although fewer serious UGI ulcer complications occurred with celecoxib than with NSAIDs, the differences between treatment groups did not achieve statistical significance (P=0.45 for celecoxib vs NSAIDs; P=0.64 for celecoxib vs diclofenac; P=0.414 for celecoxib vs ibuprofen). Patients treated with celecoxib experienced significantly fewer serious UGI ulcer complications plus symptomatic gastroduodenal ulcers than patients receiving NSAIDs (P=0.04). When compared with each NSAID, the effect of celecoxib was significant vs ibuprofen (P=0.017) and not significant vs diclofenac (P=0.296). However, since only a small fraction of ulcers are thought to result in a clinically serious outcome, symptomatic ulcers do not represent the same severity of end point as complicated perforations, ulcers, and bleeding. 1 100 200 320 100 200 300 380 Days Days Silverstein et al. JAMA. 2000;284:1247; FDA Arthritis Advisory Committee, 2001. Gaithersburg, Md. Food and Drug Administration. Celebrex: Arthritis Advisory Committee Meeting, Gaithersburg, Md; February 7, 2001. http://www.fda.gov/ohrms/dockets/ac/01/slides/3677s1.htm. Silverstein FE, Faich G, Goldstein JL, et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: the CLASS study: a randomized controlled trial. Celecoxib Long-term Arthritis Safety Study. JAMA. 2000;284:1247-1255.

塞来昔布（200-400mg/d）和非选择性NSAIDs导致的上消化道不良事件比较1 证据等级 1a 31项荟萃分析：上消化道安全性1 2006年Moore et al. 塞来昔布（200-400mg/d）和非选择性NSAIDs导致的上消化道不良事件比较1 相对风险 (95% CI) 荟萃分析了2003年10月以前，全部关于塞来昔布治疗关节炎的随机、双盲、对照研究，共计31项。 纳入患者39,605例，平均用药时间7个月。 0.71 (0.550.91) 0.70 (0.600.80) 0.0 1.0 0.35 (0.220.56) 症状性溃疡/胃肠道出血 1.25 0.75 0.50 0.25 血红蛋白减少＞20g/L 因胃肠道不耐受而停药 Moore研究荟萃分析了2003年10月以前，所有关于塞来昔布治疗关节炎的随机、双盲、对照研究，共计31项（纳入患者39,605例，患者平均用药时间7个月），旨在探讨塞来昔布治疗骨关节炎（OA）/类风湿关节炎（RA）不良事件发生率。研究证实，塞来昔布上消化道安全性显著优于非选择性NSAIDs：塞来昔布组的症状性溃疡和胃肠道出血的发生率低于非选择性NSAIDs组；血红蛋白减少的发生率、因胃肠道不耐受所致的停药率，塞来昔布组也显著低于非选择性NSAIDs组。这一结论再次证实，长期服用塞来昔布治疗OA/RA的良好的消化道耐受性及安全性。 1. McGettigan et al. JAMA. 2006. 优于非选择性NSAIDs 劣于非选择性NSAIDs 1 Moore RA, et al. Arthritis Research & Therapy 2005, 7: R644-R665

Valdecoxib vs 非选择性 NSAIDs 有显著临床意义的上消化道事件 2.5 * 1.96% 2.0 11/562 1.5 Incidence (%)† 1.0 0.68% 8/1183 0.39% 0.5 7/1791 The GI safety of valdecoxib, for doses ranging from 5 mg to 80 mg daily, was investigated in six 12- to 26-week controlled arthritis trials and 3 long-term, open label arthritis trials from 48 weeks to 15 months in duration. Ibuprofen 2400 mg daily, naproxen 1000 mg daily, and diclofenac 150 mg daily were used as comparators in the controlled arthritis trials. The combined exposure to valdecoxib during the long-term open-label study was 1352 patient years. The incidence of clinically significant UGI events was significantly lower with valdecoxib than with traditional NSAIDs and similar to that of placebo. Additional information and percentage of patients according to treatment group: Total daily dose N (%) Naproxen 1000 mg 1181 (74%) Ibuprofen 2400 mg 207 (13%) Diclofenac 150 mg 212 (13%) Valdecoxib* 3359 (5 mg—9.6%, 10 mg— 28.4%, 20 mg—25.3%, 40 mg— 24.7%, 80 mg—12.0%) *Includes pts on doses 5 mg–80 mg The incidence of clinically significant UGI events with placebo was low; however, patients did not stay on placebo for a long time before requiring active treatment. Data on file. Pharmacia Corporation. 0% 0/146 Placebo Valdecoxib NSAIDs Valdecoxib Controlled Arthritis Trials Long-Term, Open-Label Arthritis Trials *P=0.05 vs other treatments. †Events per 100 patient-years GI events = perforation, obstruction, and bleeding. Data on file. Pharmacia Corporation.

Frequency of Endoscopic Ulcers at 1 Week Healthy Patients 65 Years Placebo (n=61) Valdecoxib 40 mg bid (n=60) 20 Naproxen 500 mg bid (n=60) * † * † 15 % of Patients 10 Although there is a notion that the effects of traditional NSAIDs on the GI mucosa are insidious and thus only emerge with long-term NSAID therapy, evidence shows that GI complications can occur as early as 1 week after initiating therapy. In a study of 61 otherwise healthy elderly patients receiving double-blind therapy with the traditional NSAID naproxen 500 mg twice daily, the new COX-2–specific inhibitor valdecoxib 40 mg bid, or placebo, a significant number of patients receiving naproxen had endoscopically identified gastric and duodenal ulcers after 6 days of treatment compared with placebo. In contrast, no ulcers were identified with valdecoxib. Data on file. Pharmacia Corporation. 5 0% 0% 0% 0% Duodenal Gastric Gastroduodenal *P<0.05 vs placebo; †P<0.001 vs valdecoxib. Data on file. Pharmacia Corporation.

传统NSAIDs对小肠造成的损伤逐渐被重视 2004年6月EULAR报道： 传统NSAIDs造成不可逆的小肠损伤—小肠多发狭窄的风险 “这种隔膜从未在其他疾病中发现，仅见于传统NSAIDs相关疾病” -英国消化疾病专家Bjarnason 此隔膜只有2-3 毫米厚，放射检查未发现。停用药物后，不能吸收溶解。只能手术或肠镜才能治疗

NSAIDs and Small Intestinal Damage 10 * 8.4% 8 NSAID group (n = 249) Control group (n = 464) 6 % of Subjects With Small Intestinal Ulcers 4 2 0.6% Nonspecific ulcers *P < 0.001 Allison, et al. N Engl J Med. 1992;327:749–754.

NSAID-Induced Lesions

胶囊内窥镜试验 二级终点(mITT) 小肠黏膜损伤事件 Celecoxib (n = 115) Naproxen + omeprazole Placebo (n = 113) P value* Incidence of mucosal breaks, N (%) 18 (16%) 61 (55%) 8 (7%) < 0.001 *Across three treatments corresponding to general association of the Cochran-Mantel-Haenszel test having stratified by site

患者使用Rofecoxib 和 Naproxen后每100人年的下消化道事件发生率 Relative Risk (95% CI) 0.46(0.22-0.93) P = 0.03 1.0 0.89 0.8 0.6 Rofecoxib % of Patients with Serious Lower GI Events per 100 Patient-Years (n = 4,047) 0.41 Naproxen 0.4 (n = 4,029) 0.2 Laine, et al. Gastroenterology. 2003;124:288-292.

COX2抑制剂的问世 降低了NSAID相关性不良事件了吗? 相对胃肠道副反应减少, 但绝对数却上升 老年人NSAID的使用率由14%上升至19.8% NSAID引起的出血住院率由15.4/万人上升至17/万人 Mamdani M, et al. BMJ 2004;328:1415-6

选择性COX-2抑制剂对胃肠道的安全性只是相对的！ 大鼠急性胃黏膜损伤时有高水平的COX-2 mRNA表达， COX-2抑制剂可延缓愈合； 大鼠实验性结肠炎组织： COX-2 mRNA增加3-6倍； 敲除COX-1的大鼠： 无胃肠生理学异常， 无肉眼和镜下胃损伤， 吲哚美辛诱发的胃病变比对照组轻， 可明显减轻AA的炎症反应。 选择性COX-2抑制剂对胃肠道的安全性只是相对的！

2006年，欧洲风湿病协会（EULAR）-OA指南 对于那些对乙酰氨基酚疗效不佳的患者，应口服最低有效剂量NSAIDs（COX-2或非选择性NSAIDs ）并采用最短疗程。 对于存在胃肠高风险的患者，应该 使用选择性COX-2抑制剂 联合使用非选择性NSAIDs和胃粘膜保护剂

2006年，欧洲风湿病协会（EULAR）-OA指南 这些方案均降低NSAID相关的症状性溃疡的发生（50%-90%） 选择性COX-2抑制剂； NSAIDs+质子泵抑制剂（PPIs）； NSAIDs+H2拮抗剂； NSAIDs+米索前列醇；

2006年，欧洲风湿病协会（EULAR）-OA指南 考虑到成本效益，在预防GI事件（穿孔、溃疡或出血）成本方面，昔布类也许比合用GI保护剂更经济实惠 尽管，与非选择性NSAIDs相比，所有的方案需要额外的花费，对于那些GI出血的高危人群来说，这样花费更加符合成本效益的要求。

2007，中国骨关节炎治疗指南 NSAIDs药物治疗风险评估 上消化道不良反应高危患者 心脑肾不良反应高危患者 1 高龄(年龄>65岁) 2 长期应用 脑血管病史(有过中风史或目前有一过性脑缺血发作) 3 口服糖皮质激素 心血管病史 4 上消化道溃疡、出血病史 肾脏病史 5 使用抗凝药 同时使用血管紧张素转换酶抑制剂及利尿剂 6 酗酒史 冠脉搭桥术围手术期(禁用NSAIDs) 胃肠道高风险患者，应使用选择性COX-2抑制剂，或者使用非选择性NSAIDs加胃粘膜保护剂。

一氧化氮释放型非甾体抗炎药 主要机制： 1、一氧化氮(NO)能阻断经典NSAIDs产生的副作用，即NO在胃肠道能起到PGs相同的作用； 2、激活生物体内可溶性鸟苷酸环化酶,升高细 胞内鸟苷酸水平，从而产生多种生物效应。 拼合原理在NSAIDs上耦联一个能释放NO的部分，当药物进入体内后，立即释放出NO和NSAIDs ，NO通过抑制嗜中性粒细胞聚集，增强粘膜血流量和粘液分泌以及减少自由基生成等四个方面减少胃肠道副作用。

选择性COX-2/ LOX 双重抑制剂 生成PGs的花生四烯酸(AA)有两条代谢途径（如前所示）:①环氧合酶(COX)代谢途径，即AA在环氧合酶催化下经一系列转变代谢生成PGs；②脂氧酶(LOX)代谢途径，即AA在LOX的催化下，经一系列转变代谢生成白三烯(LTs)。 AA 的两条代谢途径中存在一定的平衡关系。即当COX的活性受到抑制时，LOX的活性增强，当LOX的活性受到抑制时，则有更多的AA进入COX代谢途径使PGs生成增加，结果都使炎症进一步发展，因而设计COX-2/LOX双重抑制剂可以达到协同消炎的目的。

NSAIDs的不良反应 胃肠道不良反应 血小板功能和出血 心血管不良反应 肾功能损伤 肝功能损害 变态反应 其它

NSAIDs和血小板功能 PGG2 and PGH2 细胞膜释放的花生四烯酸 Cox-1 阿司匹林抑制 －不可逆 阿司匹林抑制 －不可逆 非选择性NSAIDs的抑制作用－可逆 PGG2 and PGH2 内皮细胞 血小板 Another adverse consequence of non-aspirin NSAIDs is their effect on platelet function, leading to potentially serious clinical effects on hemostasis. COX-1 catalyzes the formation of arachidonic acid to prostaglandins (PG) G2 and H2. In the platelet, PGH2 is metabolized into thromboxane A2, a potent stimulus of platelet aggregation. Platelets are believed to express only COX-1 and not COX-2. In endothelial cells, PGH2 is metabolized into prostacyclin, which inhibits platelet aggregation. Traditional NSAIDs, including aspirin, which inhibit COX-1, impair the blood clotting response to GI mucosal or vascular injury. Aspirin therapy is widely recommended to reduce the risk of myocardial infarction and stroke because it irreversibly inhibits COX-1 and the formation of thromboxane A2, thus reducing platelet aggregation in response to vascular injury. The recovery of platelet function is dependent on the maturity of new platelets. In contrast, the effects of traditional NSAIDs on platelet function are reversible, so that recovery of platelet function depends on plasma concentration and clearance. The inhibition of platelet aggregation by traditional NSAIDs has important clinical implications. NSAID-related GI toxicity can be exacerbated by antiplatelet effects and potentiate the risk of GI bleeding. Similarly, patients taking traditional NSAIDs may have comorbid conditions or may be taking other medications that can increase the risk for bleeding. Since COX-2–specific inhibitors have no effect on COX-1-mediated production of thromboxane A2, they do not share the risk of platelet inhibition with traditional NSAIDs. Verburg KM, Maziasz TJ, Weiner E, et al. COX-2-specific inhibitors: definition of a new therapeutic concept. Am J Ther. 2001;8:49-64. 前列环素合成酶 血栓素合成酶 前列环素 ( 血小板聚集) TxA2 ( 血小板聚集) Verburg K, et al. Am J Ther. 2001;8:49-64.

血小板功能试验: Celecoxib vs 非选择性NSAIDs Platelet Aggregation at Steady State Placebo Celecoxib 600 mg bid Naproxen 500 mg bid Ibuprofen 800 mg tid Diclofenac 75 mg bid 20 40 60 80 100 * Platelet Aggregation (%) *† Other analyses have shown that the effect of celecoxib on platelet aggregation is similar to placebo and less than that of traditional NSAIDs. This slide shows that in Trial 1, arachidonate-induced platelet aggregation was unaffected by celecoxib or placebo administration; in contrast, platelet aggregation was almost abolished by naproxen.1 The second study confirms the lack of effect of celecoxib on platelet function compared with traditional NSAIDs demonstrated in the first study.2 Leese PT, Hubbard RC, Karim A, et al. Effects of celecoxib, a novel cyclooxygenase-2 inhibitor, on platelet function in healthy adults: a randomized, controlled trial. J Clin Pharmacol. 2000;40:124-132. Data on file. Pharmacia Corporation. *† Trial 11 (N=24) Day 10 Trial 22 (N=51) Day 8 *P<0.05 vs placebo; †P<0.05 vs celecoxib. Leese PT. J Clin Pharmacol. 2000;40:124-32. Data on file. Pharmacia Corporation.

Rofecoxib 对健康受试者血小板功能的影响 100 93.5 93.7 80 % Inhibition of Platelet Aggregation (1 mM Arachidonic-Acid Agonist) 60 40 20 This study examined whether rofecoxib would interfere with the desired antiplatelet effect of aspirin. Subjects received either rofecoxib 50 mg or placebo for 10 days. All subjects (n=12 for each group) received 81 mg aspirin daily. As expected, rofecoxib had no significant effect on serum thromboxane B2 production or platelet aggregation, further supporting the lack of effect of COX-2–specific inhibitors as a class on platelet function. Greenberg HE, Gottesdiener K, Huntington M, et al. A new cyclooxygenase-2 inhibitor, rofecoxib (Vioxx), did not alter the antiplatelet effects of low-dose aspirin in healthy volunteers. J Clin Pharmacol. 2000;40:1509–1515. 4.1 0.8 Placebo (n=12) Rofecoxib 50 mg (n=12) Aspirin 81 mg + Placebo (n=12) Aspirin 81 mg + Rofecoxib 50 mg (n=12) Day 4 Day 10 Greenberg, et al. J Clin Pharmacol. 2000;40:1509-15.

Valdecoxib vs Ibuprofen 对老年人血小板功能的影响 Platelet Aggregation With Arachidonate * * Valdecoxib 40 mg bid (n=17) Ibuprofen 800 mg tid (n=15) Placebo (n=15) * * ‡ Platelet Aggregation Response (%) † † The effects of valdecoxib on platelet studies were assessed in healthy elderly adults. Forty-seven patients were randomized to receive valdecoxib 40 mg twice daily (4 times the maximal recommended dose), ibuprofen 800 mg 3 times daily (maximal dose), or placebo for 7.5 days. The effects of ibuprofen on platelet aggregation were significantly more pronounced than those of placebo and valdecoxib. Leese PT, Recker D, Koss ME. A double-blind, placebo-controlled study to evaluate the effect of valdecoxib, a novel COX-2–specific inhibitor, on platelet function in the elderly. Ann Rheum Dis. 2001;60(suppl 1). Data on file. Pharmacia Corporation. † † 1 8 Day *P<0.001 compared with ibuprofen. †P<0.001 compared with placebo. ‡P<0.05 compared with placebo. Leese PT, et al. Ann Rheum Dis. 2001;60(suppl 1). Data on file. Pharmacia Corporation.

NSAIDs的不良反应 胃肠道不良反应 血小板功能和出血 心血管不良反应 肾功能损伤 肝功能损害 变态反应 其它 The safety profiles of the COX-2–specific inhibitors have been compared with those of traditional NSAIDs and placebo.

(Rates per 100 Patient-Years) 心血管不良事件: VIGOR Patients: RA Aspirin treatment: not allowed Patients With Events (Rates per 100 Patient-Years) Rofecoxib (n=4047) Naproxen (n=4029) Relative Risk (95% CI) Event Category Cardiac 28 (1.0) 10 (0.4) Cerebrovascular* 11 (0.4) 8 (0.3) Peripheral vascular 6 (0.2) 1 (0.0 4) 0.36 0.73 0.17 0.42 1 2 In the VIGOR study, there were 45 confirmed thrombotic events in the rofecoxib group and 19 in the naproxen group. The relative risk of sustaining a confirmed CV event on naproxen compared with rofecoxib was 0.42. The majority of events occurring in VIGOR were cardiac events. The relative risk of sustaining such an event on naproxen compared with rofecoxib was 0.36. Within the cardiac event category, most of the events were myocardial infarctions (MIs), and there was a significant reduction in MIs on naproxen compared with rofecoxib. As expected, the patients who had thrombotic events were those at higher risk—older patients, males—and ~80% of these patients had 1 or more risk factors. Confirmed CV 45 (1.7) 19 (0.7) * Not including hemorrhagic stroke. FDA Advisory Committee Meeting, 2001. Food and Drug Administration. Vioxx: Arthritis Advisory Committee Meeting, Gaithersburg, Md; February 8, 2001. http://www.fda.gov/ohrms/dockets/ac/01/slides/3677s2.htm.

不良事件 心肌梗死 Adjusted IRR of AMI or Death from CHD (95% CI) New Users During Study Reference: Nonusers IRR=1.0 Incidence Rate Ratio (IRR) Given the results of 090, ADVANTAGE, VIGOR, and the rofecoxib pooled cardiovascular analysis suggesting a cardioprotective effect of naproxen, Ray et al1 conducted a retrospective observational study to determine whether non-aspirin nonspecific NSAIDs provide cardioprotective effects; this was done by analysis of Tennessee Medicaid data that included cardiovascular risk factor status, prescription data, hospital and outpatient admissions and visits, and death certificate information. Study data revealed that by none of the rate ratios were significantly different from those of controls (1.0). The researchers concluded that even in this high-risk population of people aged 50 years, the use of nonspecific NSAIDs neither increases nor decreases the risk of serious coronary events. Thus, these data did not support the hypothesis that naproxen confers a cardioprotective benefit, nor are they consistent with the findings from 090, ADVANTAGE, VIGOR, and rofecoxib pooled cardiovascular data. As depicted in this slide, a subsequent retrospective cohort study by Ray et al2 was performed in the same database to compare the risk of acute myocardial infarction (AMI) and fatal coronary heart disease in patients receiving rofecoxib with risks in those who received other commonly prescribed NSAIDs (celecoxib, ibuprofen, and naproxen). Participants enrolled included 251,046 NSAID users (patients with an NSAID prescription or taking an NSAID during the eligibility period for the study) and 202,916 non-users (patients who had not used an NSAID within 365 days of enrollment). New users were defined as patients who began taking an individual NSAID at the beginning of the follow-up period. For inclusion, patients were required to be between 50 to 84 years of age and to have no noncardiovascular life-threatening illnesses. In terms of demographics, patients receiving COX-2 specific agents (rofecoxib and celecoxib) tended to be older and female, and to have cardiovascular or other illnesses at baseline compared with those receiving ibuprofen or naproxen. Of the new users in the study, patients who received more than 25 mg/day of rofecoxib exhibited a significantly higher incidence of serious CHD compared with those receiving other NSAID treatments, including low-dose rofecoxib (25 mg/d) (P=0.024). Compared with other COX-2 specific therapy in the study (celecoxib), the high-dose rofecoxib (25 mg/d) group exhibited 2.2 times the rate of serious CHD (P=0.014). Ray WA, Stein CM, Hall K, et al. Non-steroidal anti-inflammatory drugs and risk of serious coronary heart disease: an observational cohort study. Lancet. 2002;359:118-123. Ray WA, Stein CM, Daugherty JR, et al. COX-2 selective non-steroidal anti-inflammatory drugs and risk of serious coronary heart disease. Lancet. 2002;360:1071-1073. Ibuprofen (n=4,319) Naproxen (n=6,489) Celecoxib (n=4,509) Rofecoxib 25 mg/d (n=3,430) Rofecoxib >25 mg/d (n=500) *P=0.024 vs reference. †P=0.014 vs celecoxib (2.20 [95% CI, 1.17–4.40]). N expressed in person-years. 5,316 events. CHD = coronary heart disease. Adapted from Table 2 in Ray WA et al. Lancet. 2002;360:1071-3.

心血管不良事件: CLASS Patients: 72% with OA and 28% with RA Aspirin treatment: 21% of patients % of Patients With Events Celecoxib Diclofenac Ibuprofen Event (n=3988) (n=1996) (n=1985) MI 20 (0.5) 6 (0.3) 10 (0.5) Angina 24 (0.6) 10 (0.5) 12 (0.6) Unstable angina 12 (0.3) 4 (0.2) 2 (0.1) Any event 100 (2.5) 42 (2.1) 44 (2.2) Withdrawals 32 (0.8) 14 (0.7) 16 (0.8) As with rofecoxib, the incidence of CV AEs in the celecoxib group of CLASS was very low (<1%). The incidence of CV-related events was higher in patients taking aspirin, since they had more significant CV risk factors compared with the entire study population. However, no significant difference was observed in the incidence of CV AEs between celecoxib and the nonselective NSAID treatment groups. FDA Arthritis Advisory Committee Meeting, 2001. Food and Drug Administration. Celebrex: Arthritis Advisory Committee Meeting, Gaithersburg, Md; February 7, 2001. http://www.fda.gov/ohrms/dockets/ac/01/slides/3677s1.htm.

不良事件 心肌梗死 Crude Incidence (%) CLASS (Non-aspirin Users) 1.0 0.8 0.6 0.4 0.2 Data on myocardial infarction (MI) were collected in the CLASS study; the incidence of MI in the total population was 0.5% with celecoxib and 0.3% with traditional NSAIDs (ibuprofen and diclofenac combined) (data not shown).1 The data were further stratified by aspirin use. The subset of non-aspirin users was analyzed separately since aspirin is cardioprotective and could have altered the potential for thrombotic events with these agents. In non-aspirin users, the incidence of MI was similar between the three treatment groups. With celecoxib, the crude incidence of MI was 0.2% versus 0.1% with ibuprofen and diclofenac, evaluated either individually or combined. The differences between the treatment groups were not statistically significant. Further analyses of these data showed no increase in MI with celecoxib patients with a prior history of MI but no prophylactic aspirin use (data not shown).1 Thus, celecoxib did not increase thrombotic events in those patient at high risk for subsequent coronary events.   White WB, Falch G, Whelton A, et al. Comparison of thromboembolic events in patients treated with celecoxib, a cylooxygenase-2 specific inhibitor, versus ibuprofen or diclofenac. Am J Cardiol. 2002;89:425-430. Celecoxib (n=3,105) NSAIDs (n=3,124) CLASS = Celecoxib Long-term Arthritis Safety Study White WB, et al. Am J Cardiol. 2002;89:425-30.

Pooled Analysis from 7934 Arthritis Patients 严重血栓性心血管事件 Placebo n=1142 Pooled Analysis from 7934 Arthritis Patients Valdecoxib 10 mg (n=1543) Valdecoxib 20 mg (n=1519) Valdecoxib 40 mg (n=1066) Valdecoxib 80 mg (n=403) Incidence (%) Traditional NSAIDs n=2261 In a pooled analysis of 8,752 patients with arthritis, the incidence of serious cardiovascular events was <1.0% with valdecoxib and traditional NSAIDs, which was comparable to placebo. If the data are stratified by concurrent aspirin use, the incidence of events in aspirin users was minimal at <2.5% for all treatment groups. The slightly higher incidence of cardiovascular events in aspirin users is not surprising since many of these patients were probably receiving aspirin as a treatment for coronary artery disease, indicating an increased risk of coronary events. In non-aspirin users, the incidence of cardiovascular events in any of the treatment groups was approximately 0.5% or lower. Given these data, the risk of cardiovascular events is not increased with valdecoxib regardless of the dose administered or concomitant aspirin use.   Data on file, Pharmacia Corporation. No significant differences were observed between treatment groups. *Valdecoxib 1-80 mg daily. Data on file. Pharmacia Corporation.

Rate Ratio Coronary Heart Disease 传统NSAIDs对心血管的影响 Naproxen <1000 mg/day n=1,375† Naproxen 1000 mg/day n=3,174 Ibuprofen <1800 mg/day n=1,964 Ibuprofen 1800 mg/day n=2,994 Naproxen vs controls excluding those with previous MI/stroke‡ Rate Ratio Coronary Heart Disease * Given the results of VIGOR and the rofecoxib pooled cardiovascular analysis suggesting a cardioprotective effect of naproxen, Ray et al1 conducted a retrospective observational study to determining whether non-aspirin traditional NSAIDs provide cardioprotective effects by analyzing Tennessee Medicaid data that included cardiovascular risk factor status, prescription data, hospital and outpatient admissions and visits, and death certificate information. A cohort of 181,441 periods of new traditional NSAID use in 128,000 traditional NSAID users were compared with 181,441 periods in 134,642 matched controls. The study was conducted over an 11-year period prior to the marketing of COX-2–specific inhibitors. Several statistical analyses were performed in this study of sufficient size to test the hypothesis that naproxen has a cardioprotective effect. There were no differences in baseline characteristics, including cardiovascular risk profile, between the 2 cohorts. This slide depicts the rates of coronary heart disease associated with the use of various traditional NSAIDs used for longer than 60 uninterrupted days. None of the rate ratios were significantly different from that of controls (1.0). In addition, to assess the extent to which the use of low-dose aspirin therapy might affect findings, another rate ratio was calculated after excluding those in both cohorts with a history of MI or stroke (those for whom aspirin therapy would most likely be prescribed). This rate ratio (0.94) also did not differ from the original cohort. In summary, even in this high-risk population of people aged 50 years, the use of traditional NSAIDs neither increases nor decreases the risk of serious coronary events. Thus, these data do not support the hypothesis that naproxen confers a cardioprotective benefit, nor are they consistent with the findings from VIGOR and rofecoxib pooled cardiovascular data. Ray WA, Stein CM, Hall K, et al. Non-steroidal anti-inflammatory drugs and risk of serious coronary heart disease: an observational cohort study. Lancet. 2002;359:118-123. Use > 60 days *Control ratio = 1.0 †N expressed in person-years ‡5595 events Ray WA et al. Lancet. 2002;359:118-23.

Graham研究心血管风险同样存在于其他NSAIDs －FDA资助的回顾性研究（140万患者） 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 3.15 (1.14-8.75) P<0.01 P=0.06 P=0.01 P<0.01 P=0.005 1.69 (0.97-2.93) 校正后比值比 (95% 可信区间) 1.29 (0.93-1.79) 1.33 (1.09-1.63) 1.18 (1.04-1.35) 1.16 (1.04-1.30) 1.09 (0.99-1.21) 1.00 (对照) 0.86 (0.69-1.07) 红框中的P值均是与西乐葆相比的，与柱状图中的P值不同（此P值是与对照组，未用NSAIDs的比） 未用NSAIDs 西乐葆 布洛芬 萘普生 罗非昔布 25 mg 罗非昔布 >25 mg 其它 NSAIDs 消炎痛 双氯芬酸 AMI=急性心梗; SCD=心源性猝死.†对年龄、性别、健康计划区域、病史、吸烟情况和医疗措施应用校正。 Graham DJ, Campen D, Hui R, et al. Lancet. 2005.

COX-2选择性抑制剂与传统NSAID： 在胃肠道风险与心血管风险间的平衡 2007 Moore 荟萃研究 证据等级 1a 目的： 对比5种昔布类药物与传统NSAID的GI和CV的安全性 研究纳入标准 患者有慢性痛（RA、OA、下腰痛、强直性脊柱炎） 患者用药≥6周 随机对照试验纳入要求：双盲；评分高 观察性临床研究纳入要求：规模大；范围广 采集数据包括：事件的描述及数量；参加患者的病例数；药物剂量；患者平均暴露时间（或患者总的暴露时间）；数据源于RAM并经SD校正。 首要终点 GI终点：严重的或复杂上消化道穿孔、溃疡、出血 CV终点：APTC事件（致死或非致死性心肌梗塞、卒中、血管性死亡） 评价指标 CV和GI事件的年发生率 Ref：BMC Musculoskeletal Disorders 2007, 8:73

不同的选择性COX-2抑制剂安全性存在差异 塞来昔布 伐地昔布 罗非昔布 依托考昔 鲁米昔布 所有昔布类 昔布类药物更好 NSAIDs更好 Moore RA, et al. BMC Musculoskeletal Disorders 2007, 8:73

可能的作用机制： 选择性COX-2抑制剂抑制前列腺素但不抑制血栓素，因此会导致凝血机制中促血栓和抗血栓作用的失衡，出现促凝现象 由于减少了具有扩张作用的PGI2的产生，COX-2选择性抑制剂使得平衡向促血栓倾斜，可能增加心血管系统血栓事件的发生。

Nonspecific NSAIDs/ASA Platelet Endothelial cell Coxibs Nonspecific NSAIDs/ASA COX-1 COX-1 COX-2 血栓素 (TxA2) 前列环素 (PGI2) 血管收缩剂 促进血小板聚集 血管舒张剂 抑制血小板聚集 止血 血栓形成 McAdam et al. Proc Natl Acad Sci USA. 1999;96:272.

Coxibs: 药理学性质 吸收 Tmax = 2.8 hours T1/2 = 11.2 hours 氨苯磺胺 砜 O NH2 CF3 N S CH3 F O H3C S F Celecoxib Rofecoxib 吸收 Tmax = 2.8 hours T1/2 = 11.2 hours Steady state reached: 5 days Median Tmax = 2–3 hours T1/2 = 17 hours Steady state reached: 4 days 代谢 The potentially important pharmacologic differences between celecoxib and rofecoxib are related to chemical structure, oral bioavailability, half-life, and main pathways of hepatic metabolism. Whereas rofecoxib contains a sulfone side chain, celecoxib has a sulfonamide. This is a clinically significant difference as celecoxib is contraindicated in patients with a history of allergic reaction to sulfonamides. Rofecoxib has a significantly higher oral bioavailability compared with celecoxib (>90% vs 22%–40%, respectively). In liver via P450 2C9 60% in liver via cytosolic reductase 40% in liver via P450 3A4 动力学 Linear Nonlinear, saturable 累积 No Yes; accumulation factor: 1.67 口服生物利用度 22%–40% 92%–93% Vioxx® package insert, Merck & Co, Inc.; Celebrex® package insert, G.D. Searle & Co. Celebrex® (celecoxib). US prescribing information. G.D. Searle & Co. 2000. FitzGerald GA, Patrono C. The coxibs, selective inhibitors of cyclooxygenase-2. N Engl J Med. 2001;345:433-442. Vioxx® (rofecoxib). US product information. Merck & Co., Inc. 2000.

两种在心血管安全性上却存在显著差异的可能机制 罗非昔布作为一种砜类COX-2抑制剂，可促进脂蛋白颗粒的氧化损伤，这种潜在的致动脉粥样变的损伤却未在塞来昔布中观察到； 塞来昔布可增加NO释放，改善内皮功能，减少氧化应激和炎症反应 对碳酸酐酶(CA)的抑制； 塞来昔布——对CA有明显的抑制作用，利尿 罗非昔布——不具有这一功能 两者代谢途径的差异； 认为罗非昔布可以和醛固酮竞争胞质还原酶，会导致醛固酮增多而引起电解质、心血管等方面的表现

医生如果开出COX-2抑制剂处方，需提醒患者注意心血管不良反应的危险性 美国FDA提醒，尤其注意下列患者： －心脏搭桥术后的患者 －有冠状动脉堵塞疾病的患者包 括有胸部疼痛和心绞痛的患者 －曾有过中风或者近期有过一过 性脑缺血病史的患者 与NSAIDs一样，COX-2抑制剂也能够升高某些患者的血压，引起 心血管不了反应。因此，在开处方时，需要提醒患者，尤其对高危 患者（如老年和脱水患者）注意该不良作用。 ——Hui-Fang Cheng, Raymond C. Harris.Hypertension. 2004;43:525-530 http://www.FDA.gov/bbs/topics/news/2005/NEW01171.html

NSAIDs的不良反应 胃肠道不良反应 血小板功能和出血 心血管不良反应 肾功能损伤 肝功能损害 变态反应 其它 The safety profiles of the COX-2–specific inhibitors have been compared with those of traditional NSAIDs and placebo.

前列腺素在肾功能中的作用 花生四烯酸 COX-1, COX-2 PGE2 PGI2 刺激肾素释放  醛固酮分泌 减少 Na+  K+ 分泌 血管舒张 - GFR - 肾血流量 The 2 cyclooxygenase isoforms, COX-1 and COX-2, are main players in the formation of PGs from arachidonic acid. PGs are produced constitutively in many tissues in the body, including the brain, GI tract, and kidneys, and their synthesis is also induced at sites of inflammation. In the kidneys, PGs act as modulators of physiologic functions such as microvascular hemodynamics, tubular salt and water reabsorption, and renin release. The most important PGs in the kidneys are PGE2 and PGI2 (prostacyclin). PGE2 decreases tubular Na+ reabsorption. PGI2 stimulates renin release, which in turn stimulates secretion of aldosterone, and further leads to an increased K+ secretion at the distal nephron. PGI2 is also a potent vasodilator that preserves renal perfusion in conditions associated with decreased actual or effective circulating volume. 减少 Na+ 重吸收 GFR = glomerular filtration rate. Brater. Am J Med. 1999;107:65S. Brater DC. Effects of nonsteroidal anti-inflammatory drugs on renal function: focus on cyclooxygenase-2-selective inhibition. Am J Med. 1999;107:65S-71S.

NSAIDs 对肾功能的影响 钠潴留 花生四烯酸 PGE2 PGI2 高钾血症 急性肾衰 外周性水肿  血压  体重 COX-1 COX-2 PGE2 PGI2 钠潴留 外周性水肿  血压  体重 CHF (rarely) 高钾血症 急性肾衰 PGE2 decreases Na+ reabsorption at the cortical thick ascending limb of the loop of Henle. Thus, COX inhibition by NSAIDs leads to increased Na+ reabsorption, decreased response to diuretics (15%–20%), weight gain, occasional edema, and potentially CHF. NSAIDs can also decrease the response to antihypertensive agents and thus cause an increase in blood pressure (BP) (≤5 mmHg). PGI2 stimulates renin release and ultimately leads to an increase of K+ secretion by the distal nephron. Therapy with NSAIDs results in a hyporeninemic hypoaldosteronism that clinically manifests as type IV renal tubular acidosis and hyperkalemia. The decrease in K+ secretion can be fatal, especially in high-risk patients with renal insufficiency and diabetes. PGI2, which is also a potent vasodilator, is synthesized by the kidneys to maintain renal perfusion when a decrease of actual or effective circulating volume occurs. Administration of NSAIDs can cause sharp decreases in renal blood flow and could lead to likely acute renal failure. The risk of acute renal failure associated with NSAIDs may be dose-related. CHF = congestive heart failure. Brater. Am J Med. 1999;107:65S. Brater DC. Effects of nonsteroidal anti-inflammatory drugs on renal function: focus on cyclooxygenase-2-selective inhibition. Am J Med. 1999;107:65S-71S.

剂量依赖性下肢水肿: Rofecoxib Trials 0.0 2.0 4.0 6.0 8.0 10.0 5.4 % of patients 3.8 3.8 3.4 3.6 3.6 In the phase II/III OA studies, the incidences of lower extremity edema with 12.5 and 25 mg of rofecoxib were similar to those seen with the NSAID comparators, ibuprofen and diclofenac. In the VIGOR study, the incidence of lower-extremity edema with the 50-mg dose was higher than with 12.5 or 25 mg, or the NSAID comparator. This observation is not unexpected since these AEs are dose-related. Overall, in all treatment groups, discontinuation due to lower-extremity edema was rare, and most patients did not require a change in medication. Definition of lower-extremity edema in these studies encompasses the majority of edema-related AEs reported by investigators. Although the crude incidence rates shown here do not take into account the treatment duration, the VIGOR results were generally consistent with the results of the phase II/III OA studies. Ibuprofen 2400 mg n=847 Diclofenac 150 mg n=498 Rofecoxib 12.5 mg n=1215 Rofecoxib 25 mg n=1614 Rofecoxib 50 mg n=4047 Naproxen 1000 mg n=4029 Phase II/III OA studies1 VIGOR study2 * Investigator-reported. 1. Summary basis for approval, FDA. 2. FDA Advisory Committee Meeting, 2001. Food and Drug Administration. Vioxx: Arthritis Advisory Committee Meeting, Gaithersburg, Md; February 8, 2001. http://www.fda.gov/ohrms/dockets/ac/01/slides/3677s2.htm.

剂量依赖的外周水肿: Celecoxib Trials 10.0 8.0 † 6.0 % of patients 5.2 4.0 3.7 3.5 3.0 2.4 1.9 2.0 0.0 NSAID Comparator‡ n=1388 Celecoxib 200 mg n=1764 Celecoxib 400 mg n=1208 Celecoxib 800 mg n=3987 Diclofenac 150 mg n=1996 Ibuprofen 2400 mg n=1985 The incidence of edema in patients receiving celecoxib 200 or 400 mg/d in phase II/III OA studies and 800 mg/d in CLASS was similar to that seen with other NSAIDs (naproxen and diclofenac). In these studies, the definition of peripheral edema included both upper- and lower-extremity edema, as reported by investigators. Similar to rofecoxib, analysis of phase the II/III OA and CLASS results suggests that the incidence of peripheral edema with celecoxib is dose-dependent. Phase II/III OA studies1 CLASS2 * Peripheral edema includes both upper- and lower-extremity edema (investigator-reported). † P<0.05 vs celecoxib; ‡ Naproxen, diclofenac. 1. Summary basis for approval, FDA. 2. FDA Arthritis Advisory Committee Meeting, 1-year data, 2001. Food and Drug Administration. Celebrex: Arthritis Advisory Committee Meeting, Gaithersburg, Md; February 7, 2001. http://www.fda.gov/ohrms/dockets/ac/01/slides/3677s1.htm.

Phase II/III OA studies1 高血压*: Rofecoxib 10.0 9.7 8.0 6.0 5.5 % of patients 4.0 4.0 2.9 2.8 2.0 1.6 0.0 In the phase II/III studies of rofecoxib in patients with OA, incidence of hypertension was similar with NSAIDs and rofecoxib at 12.5 and 25 mg/d. For 50 mg/d, which is 2 times the maximum recommended dose for osteoarthritis, an expected increase in hypertension incidence was observed, in agreement with the dose-dependent effect of NSAIDs. Combined results from all these clinical trials suggest a dose-dependent effect of rofecoxib on blood pressure. Moreover, coxibs are likely to have similar effects to nonselective NSAIDs with respect to blood pressure and edema. Ibuprofen 2400 mg n=847 Diclofenac 150 mg n=498 Rofecoxib 12.5 mg n=1215 Rofecoxib 25 mg n=1614 Rofecoxib 50 mg n=4047 Naproxen 1000 mg n=4029 Phase II/III OA studies1 VIGOR study2 * Investigator-reported adverse experiences. 1. Summary basis for approval, FDA. 2. FDA Advisory Committee Meeting, 2001. Food and Drug Administration. Vioxx: Arthritis Advisory Committee Meeting, Gaithersburg, Md; February 8, 2001. http://www.fda.gov/ohrms/dockets/ac/01/slides/3677s2.htm.

Phase II/III OA studies1 高血压* : Celecoxib 4.0 3.0 3.1 % of patients 2.0 2.0 2.0 1.2 1.0 0.9 0.7 0.0 NSAID Comparator† n=1388 Celecoxib 200 mg n=1764 Celecoxib 400 mg n=1208 Celecoxib 800 mg n=3987 Diclofenac 150 mg n=1996 Ibuprofen 2400 mg n=1985 Phase II/III OA studies and CLASS demonstrate that, as with rofecoxib, there is a dose-dependent effect on the incidence of hypertension with celecoxib. These studies also show that hypertension with NSAIDs (naproxen and diclofenac) and with COX-2 selective inhibitors (celecoxib) occurs at similar rates. In conclusion, studies of rofecoxib and celecoxib vs NSAIDs suggest that renal-related AEs such as edema and hypertension are dose-dependent, mechanism-based class effects. Phase II/III OA studies1 CLASS2 * Investigator-reported. † Naproxen, diclofenac. 1. Summary basis for approval, FDA. 2. FDA Arthritis Advisory Committee Meeting, 2001. Food and Drug Administration. Celebrex: Arthritis Advisory Committee Meeting, Gaithersburg, Md; February 7, 2001. http://www.fda.gov/ohrms/dockets/ac/01/slides/3677s1.htm.

老年患者中萘普生和昔布类药物对血压的影响 7 Changes in BP on Day 14 of Treatment 6 Placebo (n=16) Naproxen 500 mg bid (n=17) Celecoxib 200 mg bid (n=17) Rofecoxib 25 mg qd (n=17) LS mean change from baseline ± SE A double-blind, placebo-controlled, parallel-group study compared the effects of equipotent doses of rofecoxib (25 mg/d), celecoxib (200 mg bid), and naproxen (500 mg bid) with placebo on BP in 67 healthy elderly patients. The study indicated that the effects of rofecoxib and celecoxib on systolic BP were not significantly different from that observed with the NSAID naproxen. Coxibs and naproxen caused a significant increase of up to 5 mmHg in the systolic BP of elderly patients after 14 days of treatment. The change induced in diastolic BP was smaller. Systolic BP Diastolic BP -3 Schwartz et al. EULAR, 2001. Abstract SAT0055. Schwartz JI, Malice MP, Lasseter KC, et al. Effect of rofecoxib, celecoxib, and naproxen on blood pressure and urinary sodium excretion in elderly volunteers. EULAR, 2001. Abstract SAT0055.

Celecoxib vs Rofecoxib WHO关于肾脏不良事件的数据 Celecoxib Rofecoxib 3.0 2.5 *† *† *† *† *† 2.0 1.5 * * IC Value * 1.0 * * 0.5 One study analyzed spontaneous reports of adverse drug reactions in the World Health Organization/Uppsala Monitoring Centre database through the 2nd quarter of 2000 to evaluate the renal adverse events associated with COX-2–specific inhibitor use. There were 2,720 adverse drug reactions reported with rofecoxib and 8,383 reported for celecoxib. Information component (IC) values were calculated using a bayesian confidence propagation neural network to evaluate disproportionality between the drug and adverse drug reactions. Both COX-2–specific inhibitors were associated with renal adverse events that are seen with traditional NSAID use; with the exception of nephritis, all IC values were significantly greater than 0, indicating a strong association. The difference between groups was significant in favor of celecoxib for water retention, abnormal renal function, acute renal failure, hypertension, and cardiac failure. Zhao SZ, Reynolds MW, Lefkowith J, et al. A comparison of renal-related adverse drug reactions between rofecoxib and celecoxib, based on events World Health Organization/Uppsala Monitoring Centre safety database. Clin Ther. 2001;23:1478-1491. 0.0 –0.5 Water Retention Abnormal Renal Function Acute Renal Failure Nephritis Hypertension Cardiac Failure *Significantly greater than zero †Significantly greater than celecoxib Zhao SZ, et al. Clin Ther. 2001;23:1478-91.

NSAIDs的不良反应 胃肠道不良反应 血小板功能和出血 心血管不良反应 肾功能损伤 肝功能损害 变态反应 其它

大多数NSAIDs可导致肝损害，从轻度的肝脏转氨酶升高到严重的肝细胞损害致死 大剂量使用保泰松可致肝损害，产生黄疸、肝炎； 长期或大剂量服用对乙酰氨基酚，常易导致严重肝毒性，以肝坏死常见，总死亡率为1％－2％。 最近新西兰药物监测中心报告COX-2抑制剂致肝毒性17例 ;另有报告11例。

对肝功能的影响 不同NSAIDs间有较大差异1 证据等级 1a 对肝功能的影响 不同NSAIDs间有较大差异1 转氨酶大于3倍正常上限 的患者的百分比 1.00 0.00 2.00 3.00 4.00 5.00 6.00 ≤13周 ＞13周 安慰剂 双氯芬酸 布洛芬 美洛昔康 塞来昔布 塞来昔布对肝功能无显著不良影响。从图表中可以看出：无论短期，还是长期服用塞来昔布，对肝功能的影响均与安慰剂相似，而短期服用双氯芬酸明显升高肝脏转氨酶，并且随着用药时间的增加，转氨酶大于3倍正常上限的患者比例也增加。 中度肝功能损害（Child-Pugh Ⅱ级）的患者，塞来昔布每日推荐剂量应减少约50%。 不推荐在重度肝功能损害的患者中使用塞来昔布。 1 Clin Gastroenterol Hepatol. 2005 May; 3(5): 489-98 Clin Gastroenterol Hepatol. 2005 May; 3(5): 489-98.

NSAIDs的不良反应 胃肠道不良反应 血小板功能和出血 心血管不良反应 肾功能损伤 肝功能损害 变态反应 其它 The safety profiles of the COX-2–specific inhibitors have been compared with those of traditional NSAIDs and placebo.

变态反应 主要表现：皮疹、荨麻疹、瘙痒、剥脱性皮炎、光敏等皮肤反应，也可见血管神经性水肿、粘膜水肿和哮喘。 哮喘症状：以阿司匹林较多见，阿司匹林性哮喘占它所引起不良反应的2/3，一般在用药后20min内出现，症状与一般哮喘相同，30－50岁中年人较易发生，女性多于男性，严重者可出现哮喘持续状态，甚至窒息死亡。

NSAIDs的不良反应 胃肠道不良反应 血小板功能和出血 心血管不良反应 肾功能损伤 肝功能损害 变态反应 其它 The safety profiles of the COX-2–specific inhibitors have been compared with those of traditional NSAIDs and placebo.

其他不良反应 Reye’s syndrome 肿瘤风险－－最近的研究表明使用阿司匹林可能会增加人类患胰腺肿瘤的风险 J Natl Cancer Inst , 2004 ,96 (1) :22－28 对骨折及软组织愈合的影响－－塞来昔布和罗非昔布可明显抑制大鼠骨折愈合 J Bone Miner Res , 2002 , 17(6) :963－976

其它不良反应 精神病事件：新西兰ADR监测中心到2002年7月共收到赛来昔布精神病事件11例；罗非昔布2例。包括精神混乱、抑郁、幻觉、焦虑以及思维异常，以女性较多，停药可恢复。 延迟或阻止排卵、受精、胚胎植等 (Lancet2001;358:1287~1289) 严重皮肤反应:注射伐地昔布时,发现多形性红斑,中毒性表皮坏死松解症

选择性cox-2抑制剂的安全性 (胃.肠.肾.子宫等) 抗炎 选择性COX-2抑制剂 COX-2 炎症部位 正常生理部位 延迟溃疡愈合 (巨噬.滑膜细胞等) 正常生理部位 (胃.肠.肾.子宫等) 不良反应 延迟溃疡愈合 加重结肠病变 诱发结肠穿孔 增加心血管事件 选择性COX-2抑制剂 抗炎

NSAIDs的药物相互作用 药动学 药效学

药动学－－蛋白结合 多数NSAIDs蛋白结合率超过95% 可能存在竞争结合位点的相互作用 华法林 阿司匹林 地高辛－吲哚美辛 This has the potential to be bad, but nsaids usually have less affinity than the competing drug and often are excreted at the same rate so that there is not actually a net plasma increase of nsaid and sometimes there is actually a decrease in serum nsaid levels due to clearance that keeps up with the displaced NSAID as it enters the blood stream.

P450相互作用 大部分P450导致的药物相互作用都会影响NSAIDs的代谢，而对与其作用的药物的代谢影响小。 What you guys are really interested in Most P450 effects are on the NSAIDs themselves, which is not exciting because fluctuation in NSAID levels aren’t as dangerous as with other drugs

CYP2C9 底物: 塞来昔布, 双氯芬酸, 双氯芬酸, 布洛芬, 吲哚美辛, 美洛昔康, 萘普生, 吡罗昔康 抑制剂: 双氯芬酸, 依托度酸*, 酮洛芬, *incredibly weak 2C9 is the CYP most involved with NSAIDs. 2C9 interactions with these drugs are most likely to affect the NSAIDs and so can be less dangerous than other combinations. Though diclofenac and ketoprofen have been shown to inhibit 2C9, there is a lack of studies that show what effect, if any, this has on the metabolism of other 2C9 substrates. Celecoxib has the most information on it because it is a newer nsaid that had great hopes of being better than the older agents.

氟康唑/伏立康唑 抑制 2C9 celecoxib的血浆浓度增加一倍 显著增高布洛芬的血药浓度 意义: NSAID血浆水平的过渡增高可能会导致肾毒性和增加心血管的危险。

利福平 诱导 2C9 显著降低celecoxib血浆浓度 镇痛作用减弱 (-) celecoxib AUC 64% Inadequate pain control may lead to use of OTC NSAIDs in addition to celecoxib which could lead to nephrotoxicity

华法林 2C9的底物 竞争代谢可能导致抗凝作用增强，celecoxib 临床试验显示对 2C9*2, *3 变异的个体可能引起出血 尽管一些NSAIDs可能抑制 2C9, 但没有显示改变华法林的药动学 Though clinical trial has shown that celecoxib has no effect on PT with warfarin, another trial has shown significant effect on individuals with specific variants.

CYP2D6 celecoxib是2D6的抑制剂 底物 临床意义? β-阻滞剂 抗抑郁药/抗精神病药物 抗组胺药 阿片类 Though celecoxib has the potential to decrease the metabolism of these drugs, it is difficult to say the clinical significance. The anti-HTN effect of beta-B can be attenuated by NSAIDs but this may antagonize the potential for increased levels with celecoxib. Increased levels of antidepressants and antipsychotics likewise may not be clinically significant because patients may not become toxic with the amount these drugs may be elevated. Ketorolac has been reported to cause hallucinations with antidepressants and has been withdrawn from the French market though that may or may not mean anything in terms of 2D6. Opiates may be given with celecoxib for increased pain control, but 2D6 is responsible for converting the opiates to their active form so rather than becoming toxic, a patient may lose pain control and/or have severe n/v esp in patients who are already poor metabolizers. There are not specific studies examining these effects.

CYP2C19 抑制剂－－吲哚美辛 底物： 卡立普多, 西酞普兰, 氯氮平, 地西泮, 多塞平，氟西汀, 苯妥英, 普萘洛尔 缺乏临床试验 Interesting note: though interactions with these drugs aren’t document in clinical trial with humans, a study with mice showed that indomethacin inhibits GABA uptake and increased psychomotor dysfunction which could be potentially dangerous in combination with the drugs listed on this slide. A clinical trial between indomethacin and men demonstrated that there was no difference in the subjective effects of diazepam vs diazepam + indomethacin.

CYP3A4 底物：美洛昔康, 双氯芬酸 抑制剂：胺碘酮, 氯霉素, 克拉霉素, 环孢霉素 A, 乙炔雌二醇, 唑类抗真菌药, 葡萄柚汁 诱导剂：巴比妥类, 卡马西平, 苯妥英, 利福平, 圣约翰草 缺乏研究 Cyclosporine has been clinically shown to significantly increase diclofenac levels which can be exceedingly dangerous because cyclosporine has been shown to increase risk of nephrotoxicity in NSAIDs even without inhibition of their respective metabolisms. Note: algorithm did predict interactions for these drugs when substrate/inhibitor/inducer info was available

药效学 抑制肾前列腺素的药物可能会影响NSAIDs的作用 增加不良反应 出血 消化道不良反应 肾毒性

抑制肾前列腺素 减弱β-阻滞剂, ACE抑制剂, 利尿剂的血压调控作用 减少甲氨蝶呤排泄，增加毒性 减少锂排泄，增加毒性 This is sort of a mixed reaction. Pharmacodynamic effects of NSAIDs interfere with the pharmacodynamic effects of anti-htns, but interfere with the pharmacokinetic properties (maybe) of lithium and methotrexate. MTX and Li are odd. There is a great deal of case reports of toxicity with NSAIDs, but not controlled trials. The mechanism isn’t exactly known, but it is possibly due to reduction in clearance of these drug due to inhibition of renal prostaglandins.

增加肾毒性 环孢霉素 A 甲氨喋呤 氨苯蝶啶 他克莫司 氨基糖苷类 All of these drugs run the risk of causing kidney damage individually. When any of them are given concurrently, the kidneys may be in extreme danger!!

增加消化道出血 SSRIs 水杨酸盐 抗凝剂 H2拮抗剂 二碳磷酸盐化合物? SSRIs apparently increase risk of GI bleed by their actions of HT3 in the stomach. H2 blockers seem to mask symptoms of GI bleed which can be exceedingly dangerous because they are often used to heal ulcers which may have been caused in conjunction with NSAID use. Bisposphonates temporarily make the stomach extremely acidic. So acidic, that patients who cannot sit or stand for at least 30 minutes after taking them are contraindicated for their use. Despite this, clinical trial has shown that patients taking bisphosphonates and nsaids are not at increased risk of gi bleed compared to those taking NSAIDS alone.

谢谢大家!