Congratulations are well deserved for this review by 27 psychiatric leaders, representing 14 countries, including 294 references¹. This highly researched, well-written paper describes the characteristics of treatment-resistant depression (TRD), including prevalence, risks, clinical features, costs, public health burden, management and treatments. Despite the wealth of information provided, lingering throughout the paper is mention of the instability and inconsistency of the TRD definition. Since the paper is about TRD, the reader is left uneasy about what to assume. In fact, the criticism of the term TRD could be a major conclusion of the review.

The authors state that “a consensus definition of TRD with predictive utility does not currently exist” and that “this is a major limitation from the viewpoints of translational research, treatment development, as well as clinical and policy decision making”¹. Comments like this permeate the paper. At first, only the reader is uneasy, but, as the paper progresses, it is clear that the authors may be as well. They conclude with many suggestions for this dilemma, which make this a landmark paper on a shifting topic.

As reviewed by the authors, the most common definition of TRD is the failure to respond to two or more antidepressants despite adequate dose, duration and adherence. This definition – the authors say – does not operationalize response, ignores partial response, does not take social functioning into account, is based on the use of standard medications, and usually does not include psychotherapy.

Quite discouragingly, the authors note that most individuals meeting the criteria for major depressive disorder with access to high-quality measurement-based care will meet the criteria for TRD. Hence, treatment resistance is one of the most commonly encountered therapeutic outcomes in persons prescribed conventional antidepressants.

Despite this pessimism, the report provides at least two suggestions for improving the situation: the inclusion of evidence-based psychotherapy and the implementation of more nuanced clinical approaches, which may improve treatment selection and patient adherence, and may even be therapeutic (what is often called the therapeutic alliance).

The authors note that, according to several studies, psychotherapy is preferred over pharmacotherapy by most people with a lived experience of depression, and, when combined with medication, facilitates coping and resilience. With this encouragement, I started to follow up on their treatment guideline references to check what has been said about psychotherapy.

Indeed, psychotherapy is included as a first line intervention in the practice guidelines for treatment of depression by both the American Psychiatric Association and the American Psychological Association^{2, 3}. Can we classify patients as resistant to treatment if the guidelines for recommended treatments are not included in the definition?

One can understand historically the reluctance to include psychotherapy in the TRD definition due to the old belief that you cannot test psychotherapy because every situation is unique. But there has been a revolution in psychotherapy development and research over the last 30 years, which has challenged that belief. Psychotherapy is now precisely defined in manuals used for training of therapists with different backgrounds. These manualized psychotherapies have been tested in numerous clinical trials in different populations and settings. The formats of treatment have evolved, and there are now individual, group and digital forms. The treatments are no longer interminable, but are time limited in frequency and duration.

Evidence-based psychotherapies for depression are now recommended by treatment guidelines in the US, Canada and Australia. In 2019, the US Preventive Task Force recommended two evidence-based psychotherapies for treatment and prevention of depression during pregnancy⁴. The World Health Organization (WHO), in its Mental Health Gap Action Plan (mhGAP) Intervention Guide, included evidence-based psychotherapy⁵. These treatments are being widely disseminated throughout the world, and recently also in low-income countries. For example, a large-scale clinical trial of interpersonal psychotherapy was carried out in Uganda⁶.

Let's glance at the substantial database of clinical trials. In 2021, a meta-analysis of efficacy, acceptability and long-term outcomes of psychotherapies was published in this journal⁷. This meta-analysis included cognitive behavioral therapy (CBT), interpersonal psychotherapy (IPT), problem-solving, behavioral activation, and non-directive supportive counseling, compared with each other or to usual care, waiting list, or pill placebos. Three hundred and thirty-one randomized clinical trials with over 34,000 patients with depression were included. A 50% reduction in symptoms was the primary outcome. The authors found that all psychotherapies were more efficacious than usual care or waiting list. There were no consistent differences between psychotherapies, with a few exceptions. The effects for most psychotherapies were still evident at one-year follow-up.

In a separate report also published in this journal⁸, a network meta-analysis of the effects of psychotherapies, pharmacotherapies and their combination for adult depression was conducted. Included were 101 clinical trials and 11,010 patients with moderate or severe major depression. In general, combined treatment was more effective than psychotherapy alone or pharmacotherapy alone in achieving response (50% reduction in symptoms) and remission. There were no significant differences between psychotherapy alone and pharmacotherapy alone. Patient found combined treatment or psychotherapy alone as more acceptable than pharmacotherapy alone.

Thus, the exclusion of evidence-based psychotherapy in the evaluation of treatment resistance may be a significant omission in the TRD definition⁹.

Let's consider the second issue raised about TRD, which is adherence to the implemented pharmacological treatments. Patients may be prescribed correct medications at proper doses, may even fill the prescriptions, but may not be taking the drugs. The authors note that 30 to 50% of patients are non-adherent to medication in the acute phase of treatment. The patient may be resistant to taking the treatment prescribed and not necessarily resistant to the treatment itself.

Accurate information to ensure adherence may revolve on the therapeutic relationship. The time spent with the patient (by the physician or a trusted team member) in a supportive manner might allow a more comprehensive assessment of the patients' symptoms, social situations surrounding the symptom onset, attitudes and knowledge, experience and fears about medications, treatment options, costs, family attitudes, lifestyle barriers, and a whole host of factors which may potentially be leading to non-adherence or non-recovery. This is not formal psychotherapy, but it can be therapeutic. The information obtained could unlock the mystery of patient resistance. What is involved may be misinformation, misunderstanding, mistrust, or mistaken treatment, rather than resistance to a treatment.

The possible addition of an evidence-based psychotherapy or the time spent to obtain a comprehensive patient evaluation may reduce the high rate of TRD. This is not a recommendation for long-term psychotherapy. Most evidence-based treatments are time-limited. It is not even a call for evidence-based psychotherapy for everyone, but it does suggest the need for a thorough evaluation and a therapeutic relationship as a beginning. Before the patient, the disease or the treatment is blamed for resistance, a therapeutic alliance and perhaps psychotherapy may be worth a try. Indeed, TRD may need psychotherapy.

代表 14 个国家的 27 名精神病学领导者（包括 294 篇参考文献）进行的这次审查是当之无愧的祝贺¹。这篇经过深入研究、写得很好的论文描述了难治性抑郁症 (TRD) 的特征，包括患病率、风险、临床特征、成本、公共卫生负担、管理和治疗。尽管提供了大量信息，但整篇论文中始终提到 TRD 定义的不稳定性和不一致之处。由于这篇论文是关于 TRD 的，因此读者对假设的内容感到不安。事实上，对TRD一词的批评可能是此次审查的一个主要结论。

作者指出，“目前不存在具有预测效用的 TRD 共识定义”，并且“从转化研究、治疗开发以及临床和政策决策的角度来看，这是一个主要限制” ¹。诸如此类的评论充斥着报纸。起初，只有读者感到不安，但随着论文的进展，很明显作者也可能会感到不安。他们针对这一困境提出了许多建议，这使得这篇论文成为关于不断变化的主题的里程碑式的论文。

正如作者所评论的，TRD 最常见的定义是尽管有足够的剂量、持续时间和依从性，但对两种或多种抗抑郁药物仍没有反应。作者说，这个定义没有具体化反应，忽略了部分反应，没有考虑社会功能，基于标准药物的使用，通常不包括心理治疗。

令人沮丧的是，作者指出，大多数符合重度抑郁症标准且能够获得高质量的基于测量的护理的个体将符合 TRD 的标准。因此，治疗抵抗是服用传统抗抑郁药的人最常见的治疗结果之一。

尽管存在这种悲观情绪，该报告至少提供了两条改善这种情况的建议：纳入循证心理治疗和实施更细致的临床方法，这可能会改善治疗选择和患者依从性，甚至可能具有治疗作用（这通常是称为治疗联盟）。

作者指出，根据几项研究，大多数有抑郁症经历的人更喜欢心理治疗，而不是药物治疗，并且与药物治疗相结合，可以促进应对和恢复能力。在这种鼓励下，我开始跟踪他们的治疗指南参考文献，以检查有关心理治疗的内容。

事实上，美国精神病学协会和美国心理学会都将心理治疗作为一线干预措施纳入抑郁症治疗实践指南^{2, 3}。如果定义中未包含推荐治疗指南，我们是否可以将患者归类为对治疗有抵抗力的患者？

人们可以理解历史上不愿意将心理治疗纳入 TRD 定义中的原因，因为人们认为每种情况都是独一无二的，因此无法测试心理治疗。但过去 30 年来，心理治疗的发展和研究发生了一场革命，挑战了这一信念。现在，心理治疗在用于培训不同背景的治疗师的手册中得到了精确的定义。这些手动心理疗法已经在不同人群和环境中进行了大量临床试验。治疗的形式已经发展，现在有个人、团体和数字形式。治疗不再是无休止的，而是在频率和持续时间上受到时间限制。

美国、加拿大和澳大利亚的治疗指南现已推荐针对抑郁症的循证心理疗法。2019 年，美国预防工作组推荐了两种循证心理疗法来治疗和预防妊娠期抑郁症⁴。世界卫生组织 (WHO) 在其心理健康差距行动计划 (mhGAP) 干预指南中纳入了循证心理治疗⁵。这些治疗方法正在世界各地广泛传播，最近也在低收入国家传播。例如，乌干达进行了大规模的人际心理治疗临床试验⁶。

让我们看一下大量的临床试验数据库。2021 年，该杂志发表了一份关于心理治疗的功效、可接受性和长期结果的荟萃分析⁷。这项荟萃分析包括认知行为疗法（CBT）、人际心理疗法（IPT）、问题解决、行为激活和非指导性支持咨询，并相互比较或与常规护理、候补名单或安慰剂进行比较。其中包括 331 项随机临床试验，涉及超过 34,000 名抑郁症患者。主要结果是症状减轻 50%。作者发现所有心理治疗都比常规护理或等候名单更有效。除了少数例外，心理治疗之间没有一致的差异。大多数心理治疗的效果在一年的随访中仍然很明显。

在该杂志上还发表的另一份报告⁸中，对心理疗法、药物疗法及其组合治疗成人抑郁症的效果进行了网络荟萃分析。其中包括 101 项临床试验和 11,010 名中度或重度重度抑郁症患者。一般来说，联合治疗比单独的心理治疗或单独的药物治疗在达到缓解（症状减轻 50%）和缓解方面更有效。单独的心理治疗和单独的药物治疗之间没有显着差异。患者发现联合治疗或单独的心理治疗比单独的药物治疗更容易接受。

因此，在治疗抵抗评估中排除基于证据的心理治疗可能是 TRD 定义中的一个重大遗漏⁹。

让我们考虑关于 TRD 的第二个问题，即坚持已实施的药物治疗。患者可能会被开出适当剂量的正确药物，甚至可能会按处方配药，但可能不会服用药物。作者指出，30% 至 50% 的患者在治疗急性期不坚持用药。患者可能对接受规定的治疗有抵抗力，但不一定对治疗本身有抵抗力。

确保依从性的准确信息可能取决于治疗关系。以支持的方式（由医生或值得信赖的团队成员）与患者共度时光，可以更全面地评估患者的症状、围绕症状发作的社会状况、态度和知识、经验以及对药物、治疗的恐惧选择、成本、家庭态度、生活方式障碍以及可能导致不坚持或无法康复的一系列因素。这不是正式的心理治疗，但它可以起到治疗作用。获得的信息可以解开患者耐药性之谜。所涉及的可能是错误信息、误解、不信任或错误治疗，而不是对治疗的抵制。

可能增加循证心理治疗或花时间获得全面的患者评估可能会降低 TRD 的高发生率。这不是长期心理治疗的建议。大多数循证治疗都是有时间限制的。它甚至不是呼吁对每个人进行基于证据的心理治疗，但它确实表明需要进行彻底的评估和治疗关系作为开始。在患者将疾病或治疗归咎于抵抗之前，治疗联盟和心理治疗可能值得一试。事实上，TRD 可能需要心理治疗。