Post by Nadica (She/Her) on Oct 7, 2024 1:52:17 GMT
作者专访丨《JAMA》最新发表HAPPEN研究成果：对慢阻肺病的呼吸治疗提出了创新性解决方案！- 2024-10-06
2024年9月16日，由首都医科大学附属北京朝阳医院（以下简称朝阳医院呼吸与危重症医学科/北京市呼吸疾病研究所作为牵头单位，联合国内30家医院协作完成的多中心研究“Effect of High-Intensity vs Low-Intensity Noninvasive Positive Pressure Ventilation on the Need for Endotracheal Intubation in Patients with an Acute Exacerbation of Chronic Obstructive Pulmonary Disease: The HAPPEN Randomized Clinical Trial”由JAMA（美国医学会杂志）在线发表。
HAPPEN研究提示，对于接受6小时传统低强度无创正压通气（NPPV）后动脉血二氧化碳分压（PaCO2）仍未降至正常的慢阻肺急性加重（AECOPD）患者，实施高强度NPPV较继续实施低强度NPPV，可显著降低患者住院期间气管插管需求率。
HAPPEN研究对慢性阻塞性肺疾病（慢阻肺病）的呼吸治疗提出了创新性解决方案，在国际上首次提出AECOPD的高强度NPPV策略并通过大规模多中心临床研究阐明该策略的有效性及安全性，研究证实了高强度NPPV在AECOPD中的临床实用价值，拓宽了高强度NPPV在临床实践中的应用指征，为临床上AECOPD的无创通气参数设置提供了重要的高级别的循证医学证据，对于进一步提升NPPV在AECOPD中的应用水平、提升普通病房AECOPD的救治能力、降低ICU入住率及其资源消耗、降低重度慢阻肺病患者的医疗经济负担将起到积极推动作用。HAPPEN研究是我国关于无创正压通气领域的研究成果首次登顶国际顶级医学期刊，标志着中国呼吸与危重症医学团队在组织实施高质量临床研究方面取得了突破性进展。
AECOPD是临床常见的呼吸危重症，也是慢阻肺病病情进展、住院次数增加、致残、致死的主要原因。大量研究证据证实传统低强度NPPV较常规氧疗可显著降低患者气管插管需求率及院内病死率，但仍有约15%的AECOPD患者面临NPPV失败、需要气管插管并接受有创通气，导致这部分患者仍然面临较高的病死风险。HAPPEN研究是一项旨在评价高强度NPPV治疗AECOPD是否优于低强度NPPV，由研究者发起的在全国30家医院呼吸非重症监护病房实施的单盲、多中心、随机对照临床试验。研究结果显示，高强度NPPV组气管插管需求率较低强度无创通气组显著降低。高强度NPPV组腹胀发生率显著高于低强度NPPV，但腹胀导致NPPV严重不耐受的比例较低强度NPPV组并未显著增加。并未观察到任何患者因腹胀拒绝NPPV以及气胸等不良反应发生。
HAPPEN研究是朝阳医院呼吸与危重症学科半个多世纪以来在学术发展上薪火相传、不断创新的成果。上世纪六十年代末，我国呼吸学科奠基人、中国工程院院士、原北京市呼吸疾病研究所所长翁心植教授抓住机遇，开展“肺心病”研究，造就了朝阳医院在慢阻肺这一研究领域的国内地位；上世纪八十年代初，呼吸与危重症医学科建立了国内较早的呼吸监护病房（RICU），开展机械通气的临床实践与系列研究，培养出一代代国内顶尖、国际知名的中国呼吸学科专业人才，并取得丰硕研究成果。
文章的第一作者为朝阳医院呼吸与危重症医学科副主任医师罗祖金，中国医学科学院阜外医院深圳医院研究员李镒冲、绵阳市第三人民医院主任医师李文军、通辽市第二人民医院主任医师李颖、北京市良乡医院主任医师乜庆荣为共同第一作者，通讯作者为朝阳医院呼吸与危重症医学科主任医师曹志新。
罗祖金：HAPPEN研究[1]是一项旨在评价高强度无创机械通气（NPPV）治疗慢阻肺急性加重（AECOPD）是否优于低强度NPPV，由研究者发起的在全国30家医院呼吸非重症监护病房实施的单盲、多中心、随机对照临床试验。研究计划纳入600例AECOPD患者，在完成300例患者的主要研究结局后，进行了一项计划中的中期分析。结果显示，主要研究结局-随机分组后的住院期间气管插管需求率达到统计学差异（单侧P=0.004），随即终止研究。300例患者中，147例被分配到高强度NPPV组并均接受高强度NPPV治疗，153例被分配到低强度NPPV组并接受低强度NPPV治疗（1例因为方案违背接受高强度NPPV治疗），所有患者均纳入主要研究结果数据分析。
①主要研究结局：高强度NPPV组气管插管需求率较低强度无创通气组显著降低(4.8% vs 13.7%; absolute difference, -9.0% [95% CI, -15.4% to -2.5%], one-sided P = .004; rate ratio, 0.35 [95%CI, 0.14-0.76])。在调整呼吸道感染诱因、从急性加重到随机化的天数、随机化时的pH值和随机化时的PaO2/FiO2后，气管插管需求率组间差异仍然存在统计学差异(adjusted rate ratio, 0.30 [95%CI, 0.11-0.69])。符合方案人群的主要结局分析和敏感性分析与主要分析一致。以年龄、性别、吸烟史和随机分组时pH值作为预先指定亚组，以及随机分组时PaCO2、随机分组时PaO2/FiO2、随机分组时呼吸频率、FEV1和APACHE II评分作为事件后特设亚组进行比较，组间治疗效果没有显著差异。
②次要研究结局：实际气管插管率组间差异无统计学意义 (3.4% [5/147] vs 3.9% [6/153]; absolute difference, -0.5% [-4.8% to 3.7%], P = .81; rate ratio, 0.87 [95%CI, 0.25-2.72]). 尽管如此，高强度组实际气管插管率显著低于低强度组实际气管插管与切换为高强度挽救性治疗后避免气管插管的联合发生率 (3.4% [5/147] vs 11.1% [17/153]; absolute difference, -7.7% [95% CI, -13.5% to -1.9%], P = .01; rate ratio, 0.31 [95%CI, 0.10-0.76]). 两组住院期间病死率及其余次要研究终点无统计学差异。
③安全性研究结局：高强度NPPV组腹胀发生率显著高于低强度NPPV(37.4% [55/147] vs 25.5% [39/153]; absolute difference, 11.9% [95% CI, 1.5% to 22.4%], P = .03; rate ratio, 1.47 [95% CI, 1.05 to 1.95])，但腹胀导致NPPV严重不耐受的比例较低强度NPPV组并未显著增加(3.4% [5/147] vs 0.7% [1/153]; absolute difference, 2.8% [95% CI, -0.7% to 7.2%], P = .12; rate ratio, 5.20 [95% CI, 0.85 to 61.85])。未观察到任何患者因腹胀拒绝NPPV以及气胸等不良反应发生。两组严重不良事件均较低。高强度NPPV组147名患者中有6名（4.1%）出现了严重碱中毒（pH>7.55），而低强度NPPV组153名患者均未出现这种情况。其他预先指定的严重不良事件在各组之间没有显著差异。
HAPPEN研究对慢阻肺病的呼吸治疗提出了创新性解决方案，在国际上首次提出AECOPD的高强度NPPV策略并通过大规模多中心临床研究阐明该策略的有效性及安全性，研究证实了高强度NPPV在AECOPD中的临床实用价值，拓宽了高强度NPPV在临床实践中的应用指征，为临床上AECOPD的无创通气参数设置提供了重要的高级别的循证医学证据，对于进一步提升NPPV在AECOPD中的应用水平、提升普通病房AECOPD的救治能力、降低ICU入住率及其资源消耗、降低重度慢阻肺病患者的医疗经济负担将起到积极推动作用。HAPPEN研究使得我国学者在无创正压通气领域的研究成果首次登顶国际顶级医学期刊，标志着中国学者在该领域组织实施高质量临床研究方面取得了突破性进展。
《呼吸界》：能够发表在医学顶刊《JAMA》，这项研究有什么“高明之处”呢？
曹志新：这项研究总结成文后，首先就投稿给《JAMA》，经过大约8个月的审稿和数次修改，最终以“原创研究（Original Investigation）”的形式发表，并配发了评论。
我们觉得，这项研究之所以能够“一投中的”，主要的原因有：
①对临床中“老生常谈”的问题，不是“熟视无睹”，而是善于发现不足之处，提炼出有待解决的重要科学问题。AECOPD是呼吸与危重症专业的常见病种，造成严重疾病负担。NPPV治疗AECOPD效果良好，可以极大降低气管插管率，并可降低病死率，这已经是近30年来的临床共识了，是临床中的“常规操作”。按理说，AECOPD的NPPV治疗在研究上已经“乏善可陈”了。其实不然，我们从临床实践和文献回顾中发现，多年以来NPPV治疗AECOPD的失败率一直维持在15%左右[2, 3]，如果能够进一步降低失败率，比如降低一半，降到7%-8%以下，应该具有重要的临床意义。
②全面了解学科发展的历史沿革和最新动态，善于和自己关切的临床问题相结合。慢阻肺缓解期的家庭NPPV一直是临床研究的热点，近20年来，以德国Wolfram Windisch医生为代表的部分欧洲学者主张以“高强度NPPV”治疗缓解期慢阻肺，并取得显著效果[4, 5]。于是，我们想“高强度NPPV”有没有可能对AECOPD也有良好效果、甚至优于现有疗法？但是，急性加重期患者的病情毕竟有别于缓解期，这一设想是否可行，需要临床研究加以验证。
③组织严密、样本量足够大、数据质量好是高水平临床多中心随机对照研究（RCT）的精髓。这之前，我们团队已经主导完成了2项NPPV治疗AECOPD的多中心（RCT）[2, 6]，在研究设计、组织、实施上取得了一定经验。经过艰苦努力，本项目可以说基本达到了“组织严密、样本量够、数据质量好”的要求，使研究的结论具有科学性和说服力，正如《JAMA》为本研究配发的评论所说：“这项研究的一个优点是在NPPV的整个应用过程中对生理和临床指标进行了全面详细的记录”[7]，这也是《JAMA》同意接受稿件的重要原因。关于多中心RCT研究的具体设计、组织、实施，罗祖金医生作为“操刀者”之一，是很有心得的。
《呼吸界》：回顾设计实施这项多中心研究的过程，有哪些重要的经验？
罗祖金 答：谈不上经验，说几点心得吧。
① 对研究背景的思考
这项研究的整个实施及完成情况一直秉承着我们的研究宗旨：从临床实践中找到问题，研究成果返回临床中去指导临床实践。在8年前的研究设计之初，在我们用NPPV治疗AECOPD过程中，有很大一部分患者在接受NPPV后，PaCO2较接受NPPV之前，不但没有降低，反而增高。在研究设计之前，我们已经在较长时间内感受到了这方面的研究困惑，发现了这样的一个临床问题。我们相信，这个问题是我们的广大同行在实际临床实践过程中可能都遇到过的。在仔细分析这个问题背后的病理生理机制后，我们发现这根本问题在于既往NPPV过程中，通气量不足，准确意义上的肺泡通气量不足所致；既往NPPV时给的潮气量不够且因面罩死腔的存在，导致NPPV时较自主呼吸时肺泡通气量不但没有增加，反而存在降低的可能。
我们再回过来头仔细阅读文献，我们有了如下总结：NPPV治疗AECOPD既往普遍采用低强度NPPV支持策略，其吸气相气道内正压（IPAP）一般低于20cmH2O，大多处于15–16cmH2O。低强度NPPV中IPAP的调节目标多为：VT6–10ml/kg，RR＜25次/min，呼吸困难症状改善，辅助呼吸肌参与改善，这属于经验性调节方式，并未得到任何临床证据支持及指南推荐。也就是说，当下没有人关注如何去调节AECOPD的NPPV参数问题。我们当时将眼光聚焦在NPPV治疗的另外一个领域时，我们又有了重要的发现，那就是高强度NPPV。高强度NPPV早先由德国学者Windisch等[8]提出，但他们当时提出这个概念，主要是应用在稳定期COPD患者上，并未用于AECOPD患者人群。我们在理论上经过长期思索，认为高强度NPPV用于AECOPD是具有理论需求性的，并具有解决临床问题的需求，从安全性上考量既然稳定期COPD患者都可以长期使用高强度NPPV，那么急性期COPD患者接受高强度NPPV也同样具有安全性成立的可能。因此，在理论及文献考量结束之后，开始小样本探索性病例研究，生理学研究[9]，随后开始着手进行多中心临床试验的研究设计。
② 对研究设计方面的考虑
研究设计是考验一项临床研究是否能还原研究初衷，达到研究目的最重要的一个环节。一项研究能够用于指导临床实践，研究设计是否贴切临床实际就成为了关键。HAPPEN研究在研究设计过程中经过多次的研究探讨，与国内外多个同行进行过多次交流，包括在study protocol[10]发表过程中，得到了编辑及审稿人很多宝贵的意见和建议；从研究设计开始到研究启动持续2年多的时间里，经过团队成员不断的细致思考及讨论论证，研究设计一直在持续不断的更新，使得这项研究的系统完整性得到了保证。
研究设计之中的一个巧妙之处，正是由王辰院士指出的，如何辨别低强度NPPV就能很好改善通气状况的这部分患者；这部分患者理论上不需要高强度NPPV，若接受高强度NPPV则面临着腹胀等不良反应的风险。我们在研究入选标准里，通过设置6小时低强度NPPV筛查期，较为妥善的解决了这个问题，同时又将面罩不耐受的这部分患者筛查出去，为后续无论是高强度还是低强度的更为顺畅的实施提供了一定的基础。
研究设计中的另外一个亮点在于，我们设计主要研究终点为气管插管需求率，并未设置为实际气管插管率或者病死率，尽管后二者是当前大多数研究经常采用的研究终点。按照设计逻辑，当低强度组患者达到气管插管标准后，可以允许其切换为高强度NPPV进行挽救性治疗，这既达到了我们设计气管插管需求率为主要研究终点的研究目的，也最大程度地保证了受试者的利益。在这个设计逻辑下，将实际气管插管率或病死率作为主要研究终点意义有限。若将实际气管插管率及病死率作为主要研究终点，则不能在达到气管插管标准时切换为高强度NPPV，这样就存在损害受试者利益的潜在风险。
③ 研究的组织实施
临床研究直接产生临床效益，同时也受困于现实临床工作的巨大干扰，这是临床研究优势，也是劣势；能否经得起临床实践的考验，组织实施极为关键，但又极为困难。HAPPEN研究的实施过程为2019年1月至2022年4月，这40个月的时间又恰逢COVID-19的侵袭，艰难程度可见一斑。研究实施前，对每个协作组单位进行study protocol的规范化培训。研究实施过程中，我们每3个月会有一次线上或线下讨论，通报研究进度，总结近期研究过程中出现的任何问题，听取各协作组单位的意见与建议；每周一都会举行微信群线上通报及线上沟通会，通报每周的研究进展。研究的每一例患者，我们都要求有独立的专业技术人员的跟踪督导，在对患者干预及观察过程中尽可能把控住每一个环节，最优化的做好质量控制。HAPPEN研究协作单位以地市级三级医院居多，这些医院条件装备较为优良、病源充分，而且参加多中心研究的热情特别高涨，同时又对研究的组织方特别信赖，有这样的研究协作组为后盾是我们的荣幸！可以说，没有这数十家协作单位的团结一致，这项研究是不可能顺利完成的！尽管经费支持极不充分，各协作单位克服一切困难，不分白昼，严格按照病例筛查标准、入组标准及操作标准，准时进行研究相关操作，准时进行研究记录，这是研究最终结局圆满的关键。
《呼吸界》：慢阻肺、呼吸衰竭历来是北京市呼吸疾病研究所的学术强项，请介绍一下本研究团队的成长历程。
曹志新 答：慢阻肺、呼吸衰竭、呼吸治疗一直是北京市呼吸疾病研究所（呼研所）的重要学术发展方向。1960年代末期，我国现代呼吸病学的奠基人之一、北京呼研所原所长、北京朝阳医院原副院长、已故翁心植院士抓住当时的历史机遇，搞“肺心病”研究[11]，取得突出成果，也造就了我们呼研所的一个重要学术发展脉络。1980年代初期，翁心植院士的学生、现中国医学科学院院长、北京协和医学院校长王辰院士在北京呼研所建立了国内较早的呼吸监护病房（RICU），引进现代机械通气治疗技术和理念，开展慢阻肺呼吸衰竭的研究。1990年代末期，王辰院士开始关注NPPV治疗AECOPD，带领我们完成了2项临床多中心RCT[6, 12]，也是当时国内规模最大的机械通气临床研究，其中一项关于以“肺部感染控制窗（PIC window）”为切换点、对AECOPD患者实施有创-无创序贯性撤机的研究，在全球范围内产生学术影响，成果被纳入国内多项临床指南。2016年，以我和罗祖金为主的团队完成了一项多中心RCT[2]，内容还是NPPV治疗AECOPD，虽然研究的样本量不是太大，发表的期刊影响力也一般，但这是我们团队首次独立设计、组织、实施临床多中心RCT，研究的顺利完成仿佛为我们注入了一针“强心剂”。从2017年起，我们团队就开始酝酿这项最终发表在《JAMA》上的研究，期间王辰院士给予了悉心指导，北京呼研所给与了组织保障，8年后这个项目也算“修成正果”吧。岁月弹指一挥间，我和罗祖金从一毕业就来到北京朝阳医院、来到北京呼研所，有幸在优秀学术文化的熏陶下成长，惟有常怀感恩之心，感恩之外惟有传承。
I apologize for the google translate. I do not speak Chinese. If you do and feel up to translating, please feel free to do so!
Exclusive interview with the author | "JAMA" recently published the HAPPEN research results: an innovative solution for the respiratory treatment of COPD!
On September 16, 2024, a multicenter study titled "Effect of High-Intensity vs Low-Intensity Noninvasive Positive Pressure Ventilation on the Need for Endotracheal Intubation in Patients with an Acute Exacerbation of Chronic Obstructive Pulmonary Disease: The HAPPEN Randomized Clinical Trial" was published online in JAMA (Journal of the American Medical Association) by Beijing Chaoyang Hospital Affiliated to Capital Medical University (hereinafter referred to as the Department of Respiratory and Critical Care Medicine/Beijing Institute of Respiratory Diseases) as the lead unit and 30 domestic hospitals in collaboration.
The HAPPEN study suggests that for patients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD) whose arterial carbon dioxide partial pressure (PaCO 2 ) has not dropped to normal after 6 hours of traditional low-intensity noninvasive positive pressure ventilation (NPPV) , implementing high-intensity NPPV can significantly reduce the rate of tracheal intubation requirements during hospitalization compared with continuing to implement low-intensity NPPV.
The HAPPEN study proposed an innovative solution for respiratory treatment of chronic obstructive pulmonary disease (COPD). It was the first time in the world to propose a high-intensity NPPV strategy for AECOPD and to clarify the effectiveness and safety of the strategy through a large-scale multicenter clinical study. The study confirmed the clinical practical value of high-intensity NPPV in AECOPD, broadened the application indications of high-intensity NPPV in clinical practice, and provided important high-level evidence-based medicine for the setting of non-invasive ventilation parameters for AECOPD in clinical practice. It will play a positive role in further improving the application level of NPPV in AECOPD, improving the treatment capacity of AECOPD in general wards, reducing ICU occupancy rate and its resource consumption, and reducing the medical economic burden of patients with severe COPD. The HAPPEN study is the first time that China's research results in the field of non-invasive positive pressure ventilation have topped the international top medical journals, marking a breakthrough in the organization and implementation of high-quality clinical research by the Chinese respiratory and critical care medicine team.
AECOPD is a common respiratory critical illness in clinical practice and the main cause of COPD progression, increased hospitalization, disability and death. A large amount of research evidence has confirmed that traditional low-intensity NPPV can significantly reduce the need for tracheal intubation and in-hospital mortality compared with conventional oxygen therapy, but about 15% of AECOPD patients still face NPPV failure, need tracheal intubation and receive invasive ventilation, resulting in these patients still facing a high risk of mortality. The HAPPEN study is a single-blind, multicenter, randomized controlled clinical trial initiated by researchers in respiratory non-intensive care units of 30 hospitals across the country to evaluate whether high-intensity NPPV is better than low-intensity NPPV in the treatment of AECOPD. The results showed that the need for tracheal intubation in the high-intensity NPPV group was significantly lower than that in the low-intensity non-invasive ventilation group. The incidence of abdominal distension in the high-intensity NPPV group was significantly higher than that in the low-intensity NPPV group, but the proportion of severe intolerance of NPPV due to abdominal distension was not significantly increased compared with the low-intensity NPPV group. No patients were observed to refuse NPPV due to abdominal distension or have adverse reactions such as pneumothorax .
The HAPPEN study is the result of the continuous innovation and academic development of the Department of Respiratory and Critical Care Medicine of Chaoyang Hospital over the past half century. In the late 1960s, Professor Weng Xinzhi, the founder of my country's respiratory discipline, academician of the Chinese Academy of Engineering, and former director of the Beijing Institute of Respiratory Diseases, seized the opportunity to carry out research on "cor pulmonale", which established Chaoyang Hospital's domestic status in the field of COPD research; in the early 1980s, the Department of Respiratory and Critical Care Medicine established one of the earliest respiratory intensive care units (RICU) in China, carried out clinical practice and series of research on mechanical ventilation, cultivated generations of top domestic and internationally renowned Chinese respiratory professionals, and achieved fruitful research results.
The first author of the article is Luo Zujin, deputy chief physician of the Department of Respiratory and Critical Care Medicine, Chaoyang Hospital. Li Yichong, researcher at the Shenzhen Hospital of the Fuwai Hospital, Chinese Academy of Medical Sciences, Li Wenjun, chief physician of Mianyang Third People's Hospital, Li Ying, chief physician of Tongliao Second People's Hospital, and Mie Qingrong, chief physician of Beijing Liangxiang Hospital, are co-first authors. The corresponding author is Cao Zhixin, chief physician of the Department of Respiratory and Critical Care Medicine, Chaoyang Hospital.
"Respiratory World": Please introduce the main academic achievements of this research.
Luo Zujin: The HAPPEN study [1] is a single-blind, multicenter, randomized controlled clinical trial initiated by researchers and conducted in respiratory non-intensive care units of 30 hospitals across the country to evaluate whether high-intensity noninvasive mechanical ventilation (NPPV) is superior to low-intensity NPPV in the treatment of acute exacerbation of chronic obstructive pulmonary disease (AECOPD). The study planned to include 600 patients with AECOPD. After completing the primary study outcomes of 300 patients, a planned interim analysis was conducted. The results showed that the primary study outcome - the rate of intubation during hospitalization after randomization - reached a statistically significant difference (one-sided P=0.004), and the study was terminated immediately. Of the 300 patients, 147 were assigned to the high-intensity NPPV group and all received high-intensity NPPV treatment, and 153 were assigned to the low-intensity NPPV group and received low-intensity NPPV treatment (1 patient received high-intensity NPPV treatment due to protocol violation). All patients were included in the data analysis of the primary study outcomes.
①Main study outcome: The rate of endotracheal intubation requirement was significantly lower in the high-intensity NPPV group than in the low-intensity noninvasive ventilation group (4.8% vs 13.7%; absolute difference, -9.0% [95% CI, -15.4% to -2.5%], one-sided P = .004; rate ratio, 0.35 [95%CI, 0.14-0.76]). After adjusting for respiratory tract infection precipitants, days from acute exacerbation to randomization, pH at randomization, and PaO 2 /FiO 2 at randomization , the difference in the rate of endotracheal intubation requirement between the two groups remained statistically significant (adjusted rate ratio, 0.30 [95%CI, 0.11-0.69]). The primary outcome analysis and sensitivity analysis of the per-protocol population were consistent with the primary analysis. There was no significant difference in treatment effect between the groups when age, sex, smoking history and pH value at randomization were used as pre-specified subgroups, and PaCO 2 at randomization, PaO 2 /FiO 2 at randomization, respiratory rate at randomization, FEV1 and APACHE II score were used as ad hoc subgroups after events.
②Secondary outcomes: There was no significant difference in the actual intubation rate between the two groups (3.4% [5/147] vs 3.9% [6/153]; absolute difference, -0.5% [-4.8% to 3.7%], P = .81; rate ratio, 0.87 [95%CI, 0.25-2.72]). However, the actual intubation rate in the high-intensity group was significantly lower than the combined incidence of actual intubation and avoidance of intubation after switching to high-intensity rescue therapy in the low-intensity group (3.4% [5/147] vs 11.1% [17/153]; absolute difference, -7.7% [95%CI, -13.5% to -1.9%], P = .01; rate ratio, 0.31 [95%CI, 0.10-0.76]). There was no significant difference in the in-hospital mortality rate and other secondary endpoints between the two groups.
③ Safety study outcomes: The incidence of abdominal distension was significantly higher in the high-intensity NPPV group than in the low-intensity NPPV group (37.4% [55/147] vs 25.5% [39/153]; absolute difference, 11.9% [95% CI, 1.5% to 22.4%], P = .03; rate ratio, 1.47 [95% CI, 1.05 to 1.95]), but the proportion of severe intolerance to NPPV due to abdominal distension was not significantly increased compared with the low-intensity NPPV group (3.4% [5/147] vs 0.7% [1/153]; absolute difference, 2.8% [95% CI, -0.7% to 7.2%], P = .12; rate ratio, 5.20 [95% CI, 0.85 to 61.85]). No patient refused NPPV due to abdominal distension or had adverse reactions such as pneumothorax. Serious adverse events were low in both groups. Severe alkalosis (pH > 7.55) occurred in 6 of 147 patients (4.1%) in the high-intensity NPPV group and in none of the 153 patients in the low-intensity NPPV group. Other prespecified serious adverse events did not differ significantly between the groups.
The HAPPEN study proposed an innovative solution for the respiratory treatment of COPD. It proposed the high-intensity NPPV strategy for AECOPD for the first time in the world and clarified the effectiveness and safety of the strategy through a large-scale multicenter clinical study. The study confirmed the clinical practical value of high-intensity NPPV in AECOPD, broadened the application indications of high-intensity NPPV in clinical practice, and provided important high-level evidence-based medicine for the setting of non-invasive ventilation parameters for AECOPD in clinical practice. It will play a positive role in further improving the application level of NPPV in AECOPD, improving the treatment capacity of AECOPD in general wards, reducing ICU occupancy rate and its resource consumption, and reducing the medical economic burden of patients with severe COPD. The HAPPEN study made the research results of Chinese scholars in the field of non-invasive positive pressure ventilation appear in the top international medical journals for the first time, marking that Chinese scholars have made breakthrough progress in organizing and implementing high-quality clinical research in this field.
"Respiratory World" : What is so "brilliant" about this research that it can be published in the top medical journal "JAMA"?
Cao Zhixin: After this research was summarized, it was first submitted to JAMA. After about 8 months of review and several revisions, it was finally published in the form of "Original Investigation" with accompanying comments.
We believe that the main reasons why this study was successful are:
① We should not turn a blind eye to the "commonplace" problems in clinical practice, but be good at discovering the shortcomings and extracting important scientific problems to be solved. AECOPD is a common disease in respiratory and critical care specialties, causing a serious disease burden. NPPV has a good effect in treating AECOPD, which can greatly reduce the rate of tracheal intubation and mortality. This has been a clinical consensus for nearly 30 years and is a "routine operation" in clinical practice. In theory, NPPV treatment of AECOPD has been "lackluster" in research. In fact, this is not the case. We have found from clinical practice and literature review that the failure rate of NPPV treatment of AECOPD has remained at around 15% for many years [2, 3] . If the failure rate can be further reduced, for example, by half to below 7%-8%, it should have important clinical significance.
② Comprehensively understand the history and latest developments of the discipline, and be good at combining it with clinical issues of concern to oneself. Family NPPV in the remission phase of COPD has always been a hot topic in clinical research. In the past 20 years, some European scholars, represented by Dr. Wolfram Windisch of Germany, have advocated the use of "high-intensity NPPV" to treat COPD in remission, and have achieved remarkable results [4, 5] . So, we wondered if "high-intensity NPPV" could also have a good effect on AECOPD, or even be better than existing therapies? However, the condition of patients in the acute exacerbation phase is different from that in the remission phase. Whether this idea is feasible needs to be verified by clinical research.
③ Tight organization, large sample size, and good data quality are the essence of high-level clinical multicenter randomized controlled trials (RCTs). Prior to this, our team had led and completed two multicenter (RCTs) of NPPV for the treatment of AECOPD [2, 6] , and gained certain experience in research design, organization, and implementation. After hard work, this project can be said to have basically met the requirements of "tight organization, large sample size, and good data quality", making the conclusions of the study scientific and convincing. As the commentary published by JAMA for this study said: "One of the advantages of this study is that the physiological and clinical indicators were fully and detailedly recorded throughout the application process of NPPV" [7] , which is also an important reason why JAMA agreed to accept the manuscript. As one of the "operators", Dr. Luo Zujin has a lot of experience in the specific design, organization, and implementation of multicenter RCT studies.
"Respiratory World" : Looking back on the process of designing and implementing this multicenter study, what important experiences have you gained?
Luo Zujin replied: I can’t really say I have experience, but I can share some of my thoughts.
① Thoughts on the research background
The entire implementation and completion of this study has always been in line with our research purpose: find problems from clinical practice, and return research results to the clinic to guide clinical practice. At the beginning of the study design 8 years ago, when we used NPPV to treat AECOPD, a large number of patients had a higher PaCO 2 after receiving NPPV than before receiving NPPV. Before the study design, we had felt the research confusion in this area for a long time and found such a clinical problem. We believe that this problem may have been encountered by our colleagues in actual clinical practice. After carefully analyzing the pathophysiological mechanism behind this problem, we found that the fundamental problem lies in the insufficient ventilation during the previous NPPV process, which is precisely caused by insufficient alveolar ventilation; the tidal volume given during the previous NPPV was insufficient and due to the existence of the mask dead space, the alveolar ventilation during NPPV did not increase, but may decrease compared with spontaneous breathing.
We went back and read the literature carefully, and we came to the following conclusion: NPPV treatment of AECOPD has generally adopted a low-intensity NPPV support strategy in the past, and its inspiratory positive airway pressure (IPAP) is generally lower than 20cmH2O , mostly at 15–16cmH2O . The adjustment targets of IPAP in low-intensity NPPV are mostly: VT6–10ml/kg, RR <25 times/min, improvement of dyspnea symptoms, and improvement of accessory respiratory muscle participation. This is an empirical adjustment method and has not been supported by any clinical evidence or recommended by guidelines. In other words, no one is currently paying attention to how to adjust the NPPV parameters for AECOPD. When we focused our attention on another area of NPPV treatment, we made another important discovery, that is, high-intensity NPPV. High-intensity NPPV was first proposed by German scholars Windisch et al . [ 8] , but when they proposed this concept, they mainly applied it to patients with stable COPD and did not use it in the AECOPD patient population. After long-term theoretical consideration, we believe that the use of high-intensity NPPV for AECOPD is theoretically necessary and has the need to solve clinical problems. From the perspective of safety, since stable COPD patients can use high-intensity NPPV for a long time, it is also possible that acute COPD patients can also receive high-intensity NPPV safely. Therefore, after completing theoretical and literature considerations, we started small sample exploratory case studies and physiological studies [9] , and then began to work on the research design of a multicenter clinical trial.
② Considerations on research design
Study design is the most important part of testing whether a clinical study can restore the original intention of the study and achieve the research purpose. Whether a study can be used to guide clinical practice, whether the study design is appropriate to clinical practice becomes the key. The HAPPEN study has undergone many research discussions during the study design process, and has had many exchanges with many domestic and foreign peers, including many valuable opinions and suggestions from editors and reviewers during the publication of the study protocol [10] . From the beginning of the study design to the start of the study, which lasted more than two years, after continuous and careful thinking and discussion by team members, the study design has been continuously updated, ensuring the system integrity of this study.
One of the clever points in the study design was pointed out by Academician Wang Chen, which is how to identify those patients whose ventilation can be improved well by low-intensity NPPV; these patients theoretically do not need high-intensity NPPV, but if they receive high-intensity NPPV, they will face the risk of adverse reactions such as abdominal distension. In the study inclusion criteria, we set up a 6-hour low-intensity NPPV screening period to solve this problem more appropriately, and at the same time screen out those patients who are intolerant to masks, providing a certain foundation for the smoother implementation of subsequent high-intensity or low-intensity treatments.
Another highlight of the study design is that we designed the primary study endpoint as the tracheal intubation requirement rate, and did not set it as the actual tracheal intubation rate or mortality rate, although the latter two are the research endpoints often used in most current studies. According to the design logic, when patients in the low-intensity group meet the tracheal intubation criteria, they can be allowed to switch to high-intensity NPPV for rescue treatment. This not only achieves the purpose of our study of setting the tracheal intubation requirement rate as the primary study endpoint, but also maximizes the interests of the subjects. Under this design logic, it is of limited significance to use the actual tracheal intubation rate or mortality rate as the primary study endpoint. If the actual tracheal intubation rate and mortality rate are used as the primary study endpoints, it is not possible to switch to high-intensity NPPV when the tracheal intubation criteria are met, which poses a potential risk of damaging the interests of the subjects.
③ Organization and implementation of the study
Clinical research directly produces clinical benefits, but it is also subject to huge interference from real clinical work. This is both an advantage and a disadvantage of clinical research; whether it can withstand the test of clinical practice, organization and implementation are extremely critical, but also extremely difficult. The implementation process of the HAPPEN study was from January 2019 to April 2022. These 40 months coincided with the invasion of COVID-19 , and the difficulty can be seen. Before the implementation of the study, standardized training on the study protocol was conducted for each collaborative group. During the implementation of the study, we will have an online or offline discussion every 3 months to report the progress of the research, summarize any problems that have arisen in the recent research process, and listen to the opinions and suggestions of each collaborative group; every Monday, we will hold an online WeChat group report and online communication meeting to report the weekly research progress. For each patient in the study, we require independent professional and technical personnel to follow up and supervise, and try to control every link in the process of patient intervention and observation, and optimize quality control. Most of the HAPPEN research collaboration units are tertiary hospitals at the prefecture and city levels. These hospitals have relatively good conditions and equipment, sufficient sources of disease, and are particularly enthusiastic about participating in multi-center research. At the same time, they have special trust in the organizers of the research. It is our honor to have such a research collaboration group as our backing! It can be said that without the unity of these dozens of collaboration units, this research would not have been successfully completed! Despite the extremely insufficient funding support, the collaboration units overcame all difficulties, worked day and night, strictly followed the case screening standards, inclusion standards and operation standards, and carried out research-related operations and research records on time. This is the key to the successful final outcome of the research.
"Respiratory World" : COPD and respiratory failure have always been the academic strengths of the Beijing Institute of Respiratory Diseases. Please introduce the growth history of this research team.
Cao Zhixin replied: COPD, respiratory failure, and respiratory therapy have always been important academic development directions of the Beijing Institute of Respiratory Diseases (BRIRD). In the late 1960s, the late Academician Weng Xinzhi, one of the founders of modern respiratory medicine in China, former director of the Beijing RIRD, former vice president of Beijing Chaoyang Hospital, seized the historical opportunity at the time to conduct research on "cor pulmonale" [11] , achieved outstanding results, and also created an important academic development context for our RIRD. In the early 1980s, Academician Wang Chen, a student of Academician Weng Xinzhi and now president of the Chinese Academy of Medical Sciences and Peking Union Medical College, established one of the earliest respiratory intensive care units (RICU) in China at the Beijing RIRD, introduced modern mechanical ventilation treatment technology and concepts, and carried out research on COPD respiratory failure. In the late 1990s, Academician Wang Chen began to pay attention to NPPV treatment for AECOPD and led us to complete two clinical multicenter RCTs [6, 12] , which were also the largest mechanical ventilation clinical studies in China at that time. One of the studies was about using the "pulmonary infection control window (PIC window)" as the switching point to implement invasive-noninvasive sequential weaning for AECOPD patients. It had an academic impact worldwide and its results were included in many clinical guidelines in China. In 2016, a team led by Luo Zujin and I completed a multicenter RCT [2] , which was also about NPPV treatment for AECOPD. Although the sample size of the study was not too large and the influence of the journal was average, it was the first time that our team independently designed, organized, and implemented a clinical multicenter RCT. The successful completion of the study seemed to have injected us with a shot of "heart booster". Since 2017, our team has begun to plan this study, which was eventually published in JAMA. During this period, Academician Wang Chen provided careful guidance and the Beijing Institute of Respiratory Diseases provided organizational support. Eight years later, this project can be considered to have "come to fruition". Time flies in the blink of an eye. Luo Zujin and I came to Beijing Chaoyang Hospital and Beijing Respiratory Research Institute as soon as we graduated. We are fortunate to grow up under the influence of excellent academic culture. We can only be grateful, and the only thing besides gratitude is inheritance.
2024年9月16日，由首都医科大学附属北京朝阳医院（以下简称朝阳医院呼吸与危重症医学科/北京市呼吸疾病研究所作为牵头单位，联合国内30家医院协作完成的多中心研究“Effect of High-Intensity vs Low-Intensity Noninvasive Positive Pressure Ventilation on the Need for Endotracheal Intubation in Patients with an Acute Exacerbation of Chronic Obstructive Pulmonary Disease: The HAPPEN Randomized Clinical Trial”由JAMA（美国医学会杂志）在线发表。
HAPPEN研究提示，对于接受6小时传统低强度无创正压通气（NPPV）后动脉血二氧化碳分压（PaCO2）仍未降至正常的慢阻肺急性加重（AECOPD）患者，实施高强度NPPV较继续实施低强度NPPV，可显著降低患者住院期间气管插管需求率。
HAPPEN研究对慢性阻塞性肺疾病（慢阻肺病）的呼吸治疗提出了创新性解决方案，在国际上首次提出AECOPD的高强度NPPV策略并通过大规模多中心临床研究阐明该策略的有效性及安全性，研究证实了高强度NPPV在AECOPD中的临床实用价值，拓宽了高强度NPPV在临床实践中的应用指征，为临床上AECOPD的无创通气参数设置提供了重要的高级别的循证医学证据，对于进一步提升NPPV在AECOPD中的应用水平、提升普通病房AECOPD的救治能力、降低ICU入住率及其资源消耗、降低重度慢阻肺病患者的医疗经济负担将起到积极推动作用。HAPPEN研究是我国关于无创正压通气领域的研究成果首次登顶国际顶级医学期刊，标志着中国呼吸与危重症医学团队在组织实施高质量临床研究方面取得了突破性进展。
AECOPD是临床常见的呼吸危重症，也是慢阻肺病病情进展、住院次数增加、致残、致死的主要原因。大量研究证据证实传统低强度NPPV较常规氧疗可显著降低患者气管插管需求率及院内病死率，但仍有约15%的AECOPD患者面临NPPV失败、需要气管插管并接受有创通气，导致这部分患者仍然面临较高的病死风险。HAPPEN研究是一项旨在评价高强度NPPV治疗AECOPD是否优于低强度NPPV，由研究者发起的在全国30家医院呼吸非重症监护病房实施的单盲、多中心、随机对照临床试验。研究结果显示，高强度NPPV组气管插管需求率较低强度无创通气组显著降低。高强度NPPV组腹胀发生率显著高于低强度NPPV，但腹胀导致NPPV严重不耐受的比例较低强度NPPV组并未显著增加。并未观察到任何患者因腹胀拒绝NPPV以及气胸等不良反应发生。
HAPPEN研究是朝阳医院呼吸与危重症学科半个多世纪以来在学术发展上薪火相传、不断创新的成果。上世纪六十年代末，我国呼吸学科奠基人、中国工程院院士、原北京市呼吸疾病研究所所长翁心植教授抓住机遇，开展“肺心病”研究，造就了朝阳医院在慢阻肺这一研究领域的国内地位；上世纪八十年代初，呼吸与危重症医学科建立了国内较早的呼吸监护病房（RICU），开展机械通气的临床实践与系列研究，培养出一代代国内顶尖、国际知名的中国呼吸学科专业人才，并取得丰硕研究成果。
文章的第一作者为朝阳医院呼吸与危重症医学科副主任医师罗祖金，中国医学科学院阜外医院深圳医院研究员李镒冲、绵阳市第三人民医院主任医师李文军、通辽市第二人民医院主任医师李颖、北京市良乡医院主任医师乜庆荣为共同第一作者，通讯作者为朝阳医院呼吸与危重症医学科主任医师曹志新。
罗祖金：HAPPEN研究[1]是一项旨在评价高强度无创机械通气（NPPV）治疗慢阻肺急性加重（AECOPD）是否优于低强度NPPV，由研究者发起的在全国30家医院呼吸非重症监护病房实施的单盲、多中心、随机对照临床试验。研究计划纳入600例AECOPD患者，在完成300例患者的主要研究结局后，进行了一项计划中的中期分析。结果显示，主要研究结局-随机分组后的住院期间气管插管需求率达到统计学差异（单侧P=0.004），随即终止研究。300例患者中，147例被分配到高强度NPPV组并均接受高强度NPPV治疗，153例被分配到低强度NPPV组并接受低强度NPPV治疗（1例因为方案违背接受高强度NPPV治疗），所有患者均纳入主要研究结果数据分析。
①主要研究结局：高强度NPPV组气管插管需求率较低强度无创通气组显著降低(4.8% vs 13.7%; absolute difference, -9.0% [95% CI, -15.4% to -2.5%], one-sided P = .004; rate ratio, 0.35 [95%CI, 0.14-0.76])。在调整呼吸道感染诱因、从急性加重到随机化的天数、随机化时的pH值和随机化时的PaO2/FiO2后，气管插管需求率组间差异仍然存在统计学差异(adjusted rate ratio, 0.30 [95%CI, 0.11-0.69])。符合方案人群的主要结局分析和敏感性分析与主要分析一致。以年龄、性别、吸烟史和随机分组时pH值作为预先指定亚组，以及随机分组时PaCO2、随机分组时PaO2/FiO2、随机分组时呼吸频率、FEV1和APACHE II评分作为事件后特设亚组进行比较，组间治疗效果没有显著差异。
②次要研究结局：实际气管插管率组间差异无统计学意义 (3.4% [5/147] vs 3.9% [6/153]; absolute difference, -0.5% [-4.8% to 3.7%], P = .81; rate ratio, 0.87 [95%CI, 0.25-2.72]). 尽管如此，高强度组实际气管插管率显著低于低强度组实际气管插管与切换为高强度挽救性治疗后避免气管插管的联合发生率 (3.4% [5/147] vs 11.1% [17/153]; absolute difference, -7.7% [95% CI, -13.5% to -1.9%], P = .01; rate ratio, 0.31 [95%CI, 0.10-0.76]). 两组住院期间病死率及其余次要研究终点无统计学差异。
③安全性研究结局：高强度NPPV组腹胀发生率显著高于低强度NPPV(37.4% [55/147] vs 25.5% [39/153]; absolute difference, 11.9% [95% CI, 1.5% to 22.4%], P = .03; rate ratio, 1.47 [95% CI, 1.05 to 1.95])，但腹胀导致NPPV严重不耐受的比例较低强度NPPV组并未显著增加(3.4% [5/147] vs 0.7% [1/153]; absolute difference, 2.8% [95% CI, -0.7% to 7.2%], P = .12; rate ratio, 5.20 [95% CI, 0.85 to 61.85])。未观察到任何患者因腹胀拒绝NPPV以及气胸等不良反应发生。两组严重不良事件均较低。高强度NPPV组147名患者中有6名（4.1%）出现了严重碱中毒（pH>7.55），而低强度NPPV组153名患者均未出现这种情况。其他预先指定的严重不良事件在各组之间没有显著差异。
HAPPEN研究对慢阻肺病的呼吸治疗提出了创新性解决方案，在国际上首次提出AECOPD的高强度NPPV策略并通过大规模多中心临床研究阐明该策略的有效性及安全性，研究证实了高强度NPPV在AECOPD中的临床实用价值，拓宽了高强度NPPV在临床实践中的应用指征，为临床上AECOPD的无创通气参数设置提供了重要的高级别的循证医学证据，对于进一步提升NPPV在AECOPD中的应用水平、提升普通病房AECOPD的救治能力、降低ICU入住率及其资源消耗、降低重度慢阻肺病患者的医疗经济负担将起到积极推动作用。HAPPEN研究使得我国学者在无创正压通气领域的研究成果首次登顶国际顶级医学期刊，标志着中国学者在该领域组织实施高质量临床研究方面取得了突破性进展。
《呼吸界》：能够发表在医学顶刊《JAMA》，这项研究有什么“高明之处”呢？
曹志新：这项研究总结成文后，首先就投稿给《JAMA》，经过大约8个月的审稿和数次修改，最终以“原创研究（Original Investigation）”的形式发表，并配发了评论。
我们觉得，这项研究之所以能够“一投中的”，主要的原因有：
①对临床中“老生常谈”的问题，不是“熟视无睹”，而是善于发现不足之处，提炼出有待解决的重要科学问题。AECOPD是呼吸与危重症专业的常见病种，造成严重疾病负担。NPPV治疗AECOPD效果良好，可以极大降低气管插管率，并可降低病死率，这已经是近30年来的临床共识了，是临床中的“常规操作”。按理说，AECOPD的NPPV治疗在研究上已经“乏善可陈”了。其实不然，我们从临床实践和文献回顾中发现，多年以来NPPV治疗AECOPD的失败率一直维持在15%左右[2, 3]，如果能够进一步降低失败率，比如降低一半，降到7%-8%以下，应该具有重要的临床意义。
②全面了解学科发展的历史沿革和最新动态，善于和自己关切的临床问题相结合。慢阻肺缓解期的家庭NPPV一直是临床研究的热点，近20年来，以德国Wolfram Windisch医生为代表的部分欧洲学者主张以“高强度NPPV”治疗缓解期慢阻肺，并取得显著效果[4, 5]。于是，我们想“高强度NPPV”有没有可能对AECOPD也有良好效果、甚至优于现有疗法？但是，急性加重期患者的病情毕竟有别于缓解期，这一设想是否可行，需要临床研究加以验证。
③组织严密、样本量足够大、数据质量好是高水平临床多中心随机对照研究（RCT）的精髓。这之前，我们团队已经主导完成了2项NPPV治疗AECOPD的多中心（RCT）[2, 6]，在研究设计、组织、实施上取得了一定经验。经过艰苦努力，本项目可以说基本达到了“组织严密、样本量够、数据质量好”的要求，使研究的结论具有科学性和说服力，正如《JAMA》为本研究配发的评论所说：“这项研究的一个优点是在NPPV的整个应用过程中对生理和临床指标进行了全面详细的记录”[7]，这也是《JAMA》同意接受稿件的重要原因。关于多中心RCT研究的具体设计、组织、实施，罗祖金医生作为“操刀者”之一，是很有心得的。
《呼吸界》：回顾设计实施这项多中心研究的过程，有哪些重要的经验？
罗祖金 答：谈不上经验，说几点心得吧。
① 对研究背景的思考
这项研究的整个实施及完成情况一直秉承着我们的研究宗旨：从临床实践中找到问题，研究成果返回临床中去指导临床实践。在8年前的研究设计之初，在我们用NPPV治疗AECOPD过程中，有很大一部分患者在接受NPPV后，PaCO2较接受NPPV之前，不但没有降低，反而增高。在研究设计之前，我们已经在较长时间内感受到了这方面的研究困惑，发现了这样的一个临床问题。我们相信，这个问题是我们的广大同行在实际临床实践过程中可能都遇到过的。在仔细分析这个问题背后的病理生理机制后，我们发现这根本问题在于既往NPPV过程中，通气量不足，准确意义上的肺泡通气量不足所致；既往NPPV时给的潮气量不够且因面罩死腔的存在，导致NPPV时较自主呼吸时肺泡通气量不但没有增加，反而存在降低的可能。
我们再回过来头仔细阅读文献，我们有了如下总结：NPPV治疗AECOPD既往普遍采用低强度NPPV支持策略，其吸气相气道内正压（IPAP）一般低于20cmH2O，大多处于15–16cmH2O。低强度NPPV中IPAP的调节目标多为：VT6–10ml/kg，RR＜25次/min，呼吸困难症状改善，辅助呼吸肌参与改善，这属于经验性调节方式，并未得到任何临床证据支持及指南推荐。也就是说，当下没有人关注如何去调节AECOPD的NPPV参数问题。我们当时将眼光聚焦在NPPV治疗的另外一个领域时，我们又有了重要的发现，那就是高强度NPPV。高强度NPPV早先由德国学者Windisch等[8]提出，但他们当时提出这个概念，主要是应用在稳定期COPD患者上，并未用于AECOPD患者人群。我们在理论上经过长期思索，认为高强度NPPV用于AECOPD是具有理论需求性的，并具有解决临床问题的需求，从安全性上考量既然稳定期COPD患者都可以长期使用高强度NPPV，那么急性期COPD患者接受高强度NPPV也同样具有安全性成立的可能。因此，在理论及文献考量结束之后，开始小样本探索性病例研究，生理学研究[9]，随后开始着手进行多中心临床试验的研究设计。
② 对研究设计方面的考虑
研究设计是考验一项临床研究是否能还原研究初衷，达到研究目的最重要的一个环节。一项研究能够用于指导临床实践，研究设计是否贴切临床实际就成为了关键。HAPPEN研究在研究设计过程中经过多次的研究探讨，与国内外多个同行进行过多次交流，包括在study protocol[10]发表过程中，得到了编辑及审稿人很多宝贵的意见和建议；从研究设计开始到研究启动持续2年多的时间里，经过团队成员不断的细致思考及讨论论证，研究设计一直在持续不断的更新，使得这项研究的系统完整性得到了保证。
研究设计之中的一个巧妙之处，正是由王辰院士指出的，如何辨别低强度NPPV就能很好改善通气状况的这部分患者；这部分患者理论上不需要高强度NPPV，若接受高强度NPPV则面临着腹胀等不良反应的风险。我们在研究入选标准里，通过设置6小时低强度NPPV筛查期，较为妥善的解决了这个问题，同时又将面罩不耐受的这部分患者筛查出去，为后续无论是高强度还是低强度的更为顺畅的实施提供了一定的基础。
研究设计中的另外一个亮点在于，我们设计主要研究终点为气管插管需求率，并未设置为实际气管插管率或者病死率，尽管后二者是当前大多数研究经常采用的研究终点。按照设计逻辑，当低强度组患者达到气管插管标准后，可以允许其切换为高强度NPPV进行挽救性治疗，这既达到了我们设计气管插管需求率为主要研究终点的研究目的，也最大程度地保证了受试者的利益。在这个设计逻辑下，将实际气管插管率或病死率作为主要研究终点意义有限。若将实际气管插管率及病死率作为主要研究终点，则不能在达到气管插管标准时切换为高强度NPPV，这样就存在损害受试者利益的潜在风险。
③ 研究的组织实施
临床研究直接产生临床效益，同时也受困于现实临床工作的巨大干扰，这是临床研究优势，也是劣势；能否经得起临床实践的考验，组织实施极为关键，但又极为困难。HAPPEN研究的实施过程为2019年1月至2022年4月，这40个月的时间又恰逢COVID-19的侵袭，艰难程度可见一斑。研究实施前，对每个协作组单位进行study protocol的规范化培训。研究实施过程中，我们每3个月会有一次线上或线下讨论，通报研究进度，总结近期研究过程中出现的任何问题，听取各协作组单位的意见与建议；每周一都会举行微信群线上通报及线上沟通会，通报每周的研究进展。研究的每一例患者，我们都要求有独立的专业技术人员的跟踪督导，在对患者干预及观察过程中尽可能把控住每一个环节，最优化的做好质量控制。HAPPEN研究协作单位以地市级三级医院居多，这些医院条件装备较为优良、病源充分，而且参加多中心研究的热情特别高涨，同时又对研究的组织方特别信赖，有这样的研究协作组为后盾是我们的荣幸！可以说，没有这数十家协作单位的团结一致，这项研究是不可能顺利完成的！尽管经费支持极不充分，各协作单位克服一切困难，不分白昼，严格按照病例筛查标准、入组标准及操作标准，准时进行研究相关操作，准时进行研究记录，这是研究最终结局圆满的关键。
《呼吸界》：慢阻肺、呼吸衰竭历来是北京市呼吸疾病研究所的学术强项，请介绍一下本研究团队的成长历程。
曹志新 答：慢阻肺、呼吸衰竭、呼吸治疗一直是北京市呼吸疾病研究所（呼研所）的重要学术发展方向。1960年代末期，我国现代呼吸病学的奠基人之一、北京呼研所原所长、北京朝阳医院原副院长、已故翁心植院士抓住当时的历史机遇，搞“肺心病”研究[11]，取得突出成果，也造就了我们呼研所的一个重要学术发展脉络。1980年代初期，翁心植院士的学生、现中国医学科学院院长、北京协和医学院校长王辰院士在北京呼研所建立了国内较早的呼吸监护病房（RICU），引进现代机械通气治疗技术和理念，开展慢阻肺呼吸衰竭的研究。1990年代末期，王辰院士开始关注NPPV治疗AECOPD，带领我们完成了2项临床多中心RCT[6, 12]，也是当时国内规模最大的机械通气临床研究，其中一项关于以“肺部感染控制窗（PIC window）”为切换点、对AECOPD患者实施有创-无创序贯性撤机的研究，在全球范围内产生学术影响，成果被纳入国内多项临床指南。2016年，以我和罗祖金为主的团队完成了一项多中心RCT[2]，内容还是NPPV治疗AECOPD，虽然研究的样本量不是太大，发表的期刊影响力也一般，但这是我们团队首次独立设计、组织、实施临床多中心RCT，研究的顺利完成仿佛为我们注入了一针“强心剂”。从2017年起，我们团队就开始酝酿这项最终发表在《JAMA》上的研究，期间王辰院士给予了悉心指导，北京呼研所给与了组织保障，8年后这个项目也算“修成正果”吧。岁月弹指一挥间，我和罗祖金从一毕业就来到北京朝阳医院、来到北京呼研所，有幸在优秀学术文化的熏陶下成长，惟有常怀感恩之心，感恩之外惟有传承。
I apologize for the google translate. I do not speak Chinese. If you do and feel up to translating, please feel free to do so!
Exclusive interview with the author | "JAMA" recently published the HAPPEN research results: an innovative solution for the respiratory treatment of COPD!
On September 16, 2024, a multicenter study titled "Effect of High-Intensity vs Low-Intensity Noninvasive Positive Pressure Ventilation on the Need for Endotracheal Intubation in Patients with an Acute Exacerbation of Chronic Obstructive Pulmonary Disease: The HAPPEN Randomized Clinical Trial" was published online in JAMA (Journal of the American Medical Association) by Beijing Chaoyang Hospital Affiliated to Capital Medical University (hereinafter referred to as the Department of Respiratory and Critical Care Medicine/Beijing Institute of Respiratory Diseases) as the lead unit and 30 domestic hospitals in collaboration.
The HAPPEN study suggests that for patients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD) whose arterial carbon dioxide partial pressure (PaCO 2 ) has not dropped to normal after 6 hours of traditional low-intensity noninvasive positive pressure ventilation (NPPV) , implementing high-intensity NPPV can significantly reduce the rate of tracheal intubation requirements during hospitalization compared with continuing to implement low-intensity NPPV.
The HAPPEN study proposed an innovative solution for respiratory treatment of chronic obstructive pulmonary disease (COPD). It was the first time in the world to propose a high-intensity NPPV strategy for AECOPD and to clarify the effectiveness and safety of the strategy through a large-scale multicenter clinical study. The study confirmed the clinical practical value of high-intensity NPPV in AECOPD, broadened the application indications of high-intensity NPPV in clinical practice, and provided important high-level evidence-based medicine for the setting of non-invasive ventilation parameters for AECOPD in clinical practice. It will play a positive role in further improving the application level of NPPV in AECOPD, improving the treatment capacity of AECOPD in general wards, reducing ICU occupancy rate and its resource consumption, and reducing the medical economic burden of patients with severe COPD. The HAPPEN study is the first time that China's research results in the field of non-invasive positive pressure ventilation have topped the international top medical journals, marking a breakthrough in the organization and implementation of high-quality clinical research by the Chinese respiratory and critical care medicine team.
AECOPD is a common respiratory critical illness in clinical practice and the main cause of COPD progression, increased hospitalization, disability and death. A large amount of research evidence has confirmed that traditional low-intensity NPPV can significantly reduce the need for tracheal intubation and in-hospital mortality compared with conventional oxygen therapy, but about 15% of AECOPD patients still face NPPV failure, need tracheal intubation and receive invasive ventilation, resulting in these patients still facing a high risk of mortality. The HAPPEN study is a single-blind, multicenter, randomized controlled clinical trial initiated by researchers in respiratory non-intensive care units of 30 hospitals across the country to evaluate whether high-intensity NPPV is better than low-intensity NPPV in the treatment of AECOPD. The results showed that the need for tracheal intubation in the high-intensity NPPV group was significantly lower than that in the low-intensity non-invasive ventilation group. The incidence of abdominal distension in the high-intensity NPPV group was significantly higher than that in the low-intensity NPPV group, but the proportion of severe intolerance of NPPV due to abdominal distension was not significantly increased compared with the low-intensity NPPV group. No patients were observed to refuse NPPV due to abdominal distension or have adverse reactions such as pneumothorax .
The HAPPEN study is the result of the continuous innovation and academic development of the Department of Respiratory and Critical Care Medicine of Chaoyang Hospital over the past half century. In the late 1960s, Professor Weng Xinzhi, the founder of my country's respiratory discipline, academician of the Chinese Academy of Engineering, and former director of the Beijing Institute of Respiratory Diseases, seized the opportunity to carry out research on "cor pulmonale", which established Chaoyang Hospital's domestic status in the field of COPD research; in the early 1980s, the Department of Respiratory and Critical Care Medicine established one of the earliest respiratory intensive care units (RICU) in China, carried out clinical practice and series of research on mechanical ventilation, cultivated generations of top domestic and internationally renowned Chinese respiratory professionals, and achieved fruitful research results.
The first author of the article is Luo Zujin, deputy chief physician of the Department of Respiratory and Critical Care Medicine, Chaoyang Hospital. Li Yichong, researcher at the Shenzhen Hospital of the Fuwai Hospital, Chinese Academy of Medical Sciences, Li Wenjun, chief physician of Mianyang Third People's Hospital, Li Ying, chief physician of Tongliao Second People's Hospital, and Mie Qingrong, chief physician of Beijing Liangxiang Hospital, are co-first authors. The corresponding author is Cao Zhixin, chief physician of the Department of Respiratory and Critical Care Medicine, Chaoyang Hospital.
"Respiratory World": Please introduce the main academic achievements of this research.
Luo Zujin: The HAPPEN study [1] is a single-blind, multicenter, randomized controlled clinical trial initiated by researchers and conducted in respiratory non-intensive care units of 30 hospitals across the country to evaluate whether high-intensity noninvasive mechanical ventilation (NPPV) is superior to low-intensity NPPV in the treatment of acute exacerbation of chronic obstructive pulmonary disease (AECOPD). The study planned to include 600 patients with AECOPD. After completing the primary study outcomes of 300 patients, a planned interim analysis was conducted. The results showed that the primary study outcome - the rate of intubation during hospitalization after randomization - reached a statistically significant difference (one-sided P=0.004), and the study was terminated immediately. Of the 300 patients, 147 were assigned to the high-intensity NPPV group and all received high-intensity NPPV treatment, and 153 were assigned to the low-intensity NPPV group and received low-intensity NPPV treatment (1 patient received high-intensity NPPV treatment due to protocol violation). All patients were included in the data analysis of the primary study outcomes.
①Main study outcome: The rate of endotracheal intubation requirement was significantly lower in the high-intensity NPPV group than in the low-intensity noninvasive ventilation group (4.8% vs 13.7%; absolute difference, -9.0% [95% CI, -15.4% to -2.5%], one-sided P = .004; rate ratio, 0.35 [95%CI, 0.14-0.76]). After adjusting for respiratory tract infection precipitants, days from acute exacerbation to randomization, pH at randomization, and PaO 2 /FiO 2 at randomization , the difference in the rate of endotracheal intubation requirement between the two groups remained statistically significant (adjusted rate ratio, 0.30 [95%CI, 0.11-0.69]). The primary outcome analysis and sensitivity analysis of the per-protocol population were consistent with the primary analysis. There was no significant difference in treatment effect between the groups when age, sex, smoking history and pH value at randomization were used as pre-specified subgroups, and PaCO 2 at randomization, PaO 2 /FiO 2 at randomization, respiratory rate at randomization, FEV1 and APACHE II score were used as ad hoc subgroups after events.
②Secondary outcomes: There was no significant difference in the actual intubation rate between the two groups (3.4% [5/147] vs 3.9% [6/153]; absolute difference, -0.5% [-4.8% to 3.7%], P = .81; rate ratio, 0.87 [95%CI, 0.25-2.72]). However, the actual intubation rate in the high-intensity group was significantly lower than the combined incidence of actual intubation and avoidance of intubation after switching to high-intensity rescue therapy in the low-intensity group (3.4% [5/147] vs 11.1% [17/153]; absolute difference, -7.7% [95%CI, -13.5% to -1.9%], P = .01; rate ratio, 0.31 [95%CI, 0.10-0.76]). There was no significant difference in the in-hospital mortality rate and other secondary endpoints between the two groups.
③ Safety study outcomes: The incidence of abdominal distension was significantly higher in the high-intensity NPPV group than in the low-intensity NPPV group (37.4% [55/147] vs 25.5% [39/153]; absolute difference, 11.9% [95% CI, 1.5% to 22.4%], P = .03; rate ratio, 1.47 [95% CI, 1.05 to 1.95]), but the proportion of severe intolerance to NPPV due to abdominal distension was not significantly increased compared with the low-intensity NPPV group (3.4% [5/147] vs 0.7% [1/153]; absolute difference, 2.8% [95% CI, -0.7% to 7.2%], P = .12; rate ratio, 5.20 [95% CI, 0.85 to 61.85]). No patient refused NPPV due to abdominal distension or had adverse reactions such as pneumothorax. Serious adverse events were low in both groups. Severe alkalosis (pH > 7.55) occurred in 6 of 147 patients (4.1%) in the high-intensity NPPV group and in none of the 153 patients in the low-intensity NPPV group. Other prespecified serious adverse events did not differ significantly between the groups.
The HAPPEN study proposed an innovative solution for the respiratory treatment of COPD. It proposed the high-intensity NPPV strategy for AECOPD for the first time in the world and clarified the effectiveness and safety of the strategy through a large-scale multicenter clinical study. The study confirmed the clinical practical value of high-intensity NPPV in AECOPD, broadened the application indications of high-intensity NPPV in clinical practice, and provided important high-level evidence-based medicine for the setting of non-invasive ventilation parameters for AECOPD in clinical practice. It will play a positive role in further improving the application level of NPPV in AECOPD, improving the treatment capacity of AECOPD in general wards, reducing ICU occupancy rate and its resource consumption, and reducing the medical economic burden of patients with severe COPD. The HAPPEN study made the research results of Chinese scholars in the field of non-invasive positive pressure ventilation appear in the top international medical journals for the first time, marking that Chinese scholars have made breakthrough progress in organizing and implementing high-quality clinical research in this field.
"Respiratory World" : What is so "brilliant" about this research that it can be published in the top medical journal "JAMA"?
Cao Zhixin: After this research was summarized, it was first submitted to JAMA. After about 8 months of review and several revisions, it was finally published in the form of "Original Investigation" with accompanying comments.
We believe that the main reasons why this study was successful are:
① We should not turn a blind eye to the "commonplace" problems in clinical practice, but be good at discovering the shortcomings and extracting important scientific problems to be solved. AECOPD is a common disease in respiratory and critical care specialties, causing a serious disease burden. NPPV has a good effect in treating AECOPD, which can greatly reduce the rate of tracheal intubation and mortality. This has been a clinical consensus for nearly 30 years and is a "routine operation" in clinical practice. In theory, NPPV treatment of AECOPD has been "lackluster" in research. In fact, this is not the case. We have found from clinical practice and literature review that the failure rate of NPPV treatment of AECOPD has remained at around 15% for many years [2, 3] . If the failure rate can be further reduced, for example, by half to below 7%-8%, it should have important clinical significance.
② Comprehensively understand the history and latest developments of the discipline, and be good at combining it with clinical issues of concern to oneself. Family NPPV in the remission phase of COPD has always been a hot topic in clinical research. In the past 20 years, some European scholars, represented by Dr. Wolfram Windisch of Germany, have advocated the use of "high-intensity NPPV" to treat COPD in remission, and have achieved remarkable results [4, 5] . So, we wondered if "high-intensity NPPV" could also have a good effect on AECOPD, or even be better than existing therapies? However, the condition of patients in the acute exacerbation phase is different from that in the remission phase. Whether this idea is feasible needs to be verified by clinical research.
③ Tight organization, large sample size, and good data quality are the essence of high-level clinical multicenter randomized controlled trials (RCTs). Prior to this, our team had led and completed two multicenter (RCTs) of NPPV for the treatment of AECOPD [2, 6] , and gained certain experience in research design, organization, and implementation. After hard work, this project can be said to have basically met the requirements of "tight organization, large sample size, and good data quality", making the conclusions of the study scientific and convincing. As the commentary published by JAMA for this study said: "One of the advantages of this study is that the physiological and clinical indicators were fully and detailedly recorded throughout the application process of NPPV" [7] , which is also an important reason why JAMA agreed to accept the manuscript. As one of the "operators", Dr. Luo Zujin has a lot of experience in the specific design, organization, and implementation of multicenter RCT studies.
"Respiratory World" : Looking back on the process of designing and implementing this multicenter study, what important experiences have you gained?
Luo Zujin replied: I can’t really say I have experience, but I can share some of my thoughts.
① Thoughts on the research background
The entire implementation and completion of this study has always been in line with our research purpose: find problems from clinical practice, and return research results to the clinic to guide clinical practice. At the beginning of the study design 8 years ago, when we used NPPV to treat AECOPD, a large number of patients had a higher PaCO 2 after receiving NPPV than before receiving NPPV. Before the study design, we had felt the research confusion in this area for a long time and found such a clinical problem. We believe that this problem may have been encountered by our colleagues in actual clinical practice. After carefully analyzing the pathophysiological mechanism behind this problem, we found that the fundamental problem lies in the insufficient ventilation during the previous NPPV process, which is precisely caused by insufficient alveolar ventilation; the tidal volume given during the previous NPPV was insufficient and due to the existence of the mask dead space, the alveolar ventilation during NPPV did not increase, but may decrease compared with spontaneous breathing.
We went back and read the literature carefully, and we came to the following conclusion: NPPV treatment of AECOPD has generally adopted a low-intensity NPPV support strategy in the past, and its inspiratory positive airway pressure (IPAP) is generally lower than 20cmH2O , mostly at 15–16cmH2O . The adjustment targets of IPAP in low-intensity NPPV are mostly: VT6–10ml/kg, RR <25 times/min, improvement of dyspnea symptoms, and improvement of accessory respiratory muscle participation. This is an empirical adjustment method and has not been supported by any clinical evidence or recommended by guidelines. In other words, no one is currently paying attention to how to adjust the NPPV parameters for AECOPD. When we focused our attention on another area of NPPV treatment, we made another important discovery, that is, high-intensity NPPV. High-intensity NPPV was first proposed by German scholars Windisch et al . [ 8] , but when they proposed this concept, they mainly applied it to patients with stable COPD and did not use it in the AECOPD patient population. After long-term theoretical consideration, we believe that the use of high-intensity NPPV for AECOPD is theoretically necessary and has the need to solve clinical problems. From the perspective of safety, since stable COPD patients can use high-intensity NPPV for a long time, it is also possible that acute COPD patients can also receive high-intensity NPPV safely. Therefore, after completing theoretical and literature considerations, we started small sample exploratory case studies and physiological studies [9] , and then began to work on the research design of a multicenter clinical trial.
② Considerations on research design
Study design is the most important part of testing whether a clinical study can restore the original intention of the study and achieve the research purpose. Whether a study can be used to guide clinical practice, whether the study design is appropriate to clinical practice becomes the key. The HAPPEN study has undergone many research discussions during the study design process, and has had many exchanges with many domestic and foreign peers, including many valuable opinions and suggestions from editors and reviewers during the publication of the study protocol [10] . From the beginning of the study design to the start of the study, which lasted more than two years, after continuous and careful thinking and discussion by team members, the study design has been continuously updated, ensuring the system integrity of this study.
One of the clever points in the study design was pointed out by Academician Wang Chen, which is how to identify those patients whose ventilation can be improved well by low-intensity NPPV; these patients theoretically do not need high-intensity NPPV, but if they receive high-intensity NPPV, they will face the risk of adverse reactions such as abdominal distension. In the study inclusion criteria, we set up a 6-hour low-intensity NPPV screening period to solve this problem more appropriately, and at the same time screen out those patients who are intolerant to masks, providing a certain foundation for the smoother implementation of subsequent high-intensity or low-intensity treatments.
Another highlight of the study design is that we designed the primary study endpoint as the tracheal intubation requirement rate, and did not set it as the actual tracheal intubation rate or mortality rate, although the latter two are the research endpoints often used in most current studies. According to the design logic, when patients in the low-intensity group meet the tracheal intubation criteria, they can be allowed to switch to high-intensity NPPV for rescue treatment. This not only achieves the purpose of our study of setting the tracheal intubation requirement rate as the primary study endpoint, but also maximizes the interests of the subjects. Under this design logic, it is of limited significance to use the actual tracheal intubation rate or mortality rate as the primary study endpoint. If the actual tracheal intubation rate and mortality rate are used as the primary study endpoints, it is not possible to switch to high-intensity NPPV when the tracheal intubation criteria are met, which poses a potential risk of damaging the interests of the subjects.
③ Organization and implementation of the study
Clinical research directly produces clinical benefits, but it is also subject to huge interference from real clinical work. This is both an advantage and a disadvantage of clinical research; whether it can withstand the test of clinical practice, organization and implementation are extremely critical, but also extremely difficult. The implementation process of the HAPPEN study was from January 2019 to April 2022. These 40 months coincided with the invasion of COVID-19 , and the difficulty can be seen. Before the implementation of the study, standardized training on the study protocol was conducted for each collaborative group. During the implementation of the study, we will have an online or offline discussion every 3 months to report the progress of the research, summarize any problems that have arisen in the recent research process, and listen to the opinions and suggestions of each collaborative group; every Monday, we will hold an online WeChat group report and online communication meeting to report the weekly research progress. For each patient in the study, we require independent professional and technical personnel to follow up and supervise, and try to control every link in the process of patient intervention and observation, and optimize quality control. Most of the HAPPEN research collaboration units are tertiary hospitals at the prefecture and city levels. These hospitals have relatively good conditions and equipment, sufficient sources of disease, and are particularly enthusiastic about participating in multi-center research. At the same time, they have special trust in the organizers of the research. It is our honor to have such a research collaboration group as our backing! It can be said that without the unity of these dozens of collaboration units, this research would not have been successfully completed! Despite the extremely insufficient funding support, the collaboration units overcame all difficulties, worked day and night, strictly followed the case screening standards, inclusion standards and operation standards, and carried out research-related operations and research records on time. This is the key to the successful final outcome of the research.
"Respiratory World" : COPD and respiratory failure have always been the academic strengths of the Beijing Institute of Respiratory Diseases. Please introduce the growth history of this research team.
Cao Zhixin replied: COPD, respiratory failure, and respiratory therapy have always been important academic development directions of the Beijing Institute of Respiratory Diseases (BRIRD). In the late 1960s, the late Academician Weng Xinzhi, one of the founders of modern respiratory medicine in China, former director of the Beijing RIRD, former vice president of Beijing Chaoyang Hospital, seized the historical opportunity at the time to conduct research on "cor pulmonale" [11] , achieved outstanding results, and also created an important academic development context for our RIRD. In the early 1980s, Academician Wang Chen, a student of Academician Weng Xinzhi and now president of the Chinese Academy of Medical Sciences and Peking Union Medical College, established one of the earliest respiratory intensive care units (RICU) in China at the Beijing RIRD, introduced modern mechanical ventilation treatment technology and concepts, and carried out research on COPD respiratory failure. In the late 1990s, Academician Wang Chen began to pay attention to NPPV treatment for AECOPD and led us to complete two clinical multicenter RCTs [6, 12] , which were also the largest mechanical ventilation clinical studies in China at that time. One of the studies was about using the "pulmonary infection control window (PIC window)" as the switching point to implement invasive-noninvasive sequential weaning for AECOPD patients. It had an academic impact worldwide and its results were included in many clinical guidelines in China. In 2016, a team led by Luo Zujin and I completed a multicenter RCT [2] , which was also about NPPV treatment for AECOPD. Although the sample size of the study was not too large and the influence of the journal was average, it was the first time that our team independently designed, organized, and implemented a clinical multicenter RCT. The successful completion of the study seemed to have injected us with a shot of "heart booster". Since 2017, our team has begun to plan this study, which was eventually published in JAMA. During this period, Academician Wang Chen provided careful guidance and the Beijing Institute of Respiratory Diseases provided organizational support. Eight years later, this project can be considered to have "come to fruition". Time flies in the blink of an eye. Luo Zujin and I came to Beijing Chaoyang Hospital and Beijing Respiratory Research Institute as soon as we graduated. We are fortunate to grow up under the influence of excellent academic culture. We can only be grateful, and the only thing besides gratitude is inheritance.