文献素材介绍
今天我们用2023年发表在 BMC Surgery 上的一篇论文作为反面教材，学习这论文的组织，写什么不写什么，需要交代的事情具体放在哪个位置。
Background
The evidence of breast‐conserving therapy (BCT) applied in centrally located breast cancer (CLBC) is absent. This study aims to investigate the long‐term survival of breast‐conserving therapy (BCT) in centrally located breast cancer (CLBC) compared with mastectomy in CLBC and BCT in non‐CLBC.
Methods
Two hundred ten thousand four hundred nine women with unilateral T1‐2 breast cancer undergoing BCT or mastectomy were identified from the Surveillance, Epidemiology, and End Results database.Kaplan–Meier survival curves were assessed via log‐rank test. Propensity score matching (PSM) was used to balance baseline features, and the multivariable Cox model was used to estimate the adjusted hazard ratio [HR] and its 95% confidence interval [CI] for breast cancer‐specific survival (BCSS) and overall survival (OS).
Methods
With a median follow‐up of 91 months, the BCSS and OS rates in patients who received BCT were greater than those patients treated with mastectomy in the entire CLBC set.
01
问题一
在所谓CLBC数据集，CLBC – centrally located breast cancer，多少个人？有人可能会说，上面方法部分中不是说了么？21万4百零九个！
错！那是unilateral T1‐2 breast cancer undergoing BCT or mastectomy 的患者，没说是中央型的，那就可以默认既包括中央型的也包括外周型的。这个道理没错吧？？
到底怎么回事情？我进入正文，看到这样流程图，的确是证实了我的猜测。
作者在上面讲到的21万4百零九个病人中间，只有不到10%，1万5千13人，得的是中央型乳腺癌。回到论文标题，讲的是centrally located breast cancer 中央型乳腺癌，结果对样本量的描述根本不是这个东西，牛头不对马嘴。
问题二
02
“中央型乳腺癌，接受保乳手术患者和常规手术相比，BCSS 和 OS 都更高”，我不信：你要让我相信，就必须给出BCSS 和 OS在2个组别中分别是多少。没有数字，说谁高谁低，那就是耍流氓。
Results
Multivariable Cox analyses showed that CLBC patients who received BCT had better BCSS (HR = 0.67, 95%CI: 0.55–0.80, p < 0.001) and OS (HR = 0.78, 95%CI: 0.68– 0.90, p = 0.001) than patients who received a mastectomy, but there were no significant differences of BCSS (HR = 0.65, 95%CI: 0.47–0.90, p = 0.009) and OS (HR = 0.82, 95%CI: 0.65–1.04, p = 0.110) after PSM.In patients treated with BCT, CLBC patients had a similar BCSS (HR = 0.99, 95%CI: 0.87–1.12, p = 0.850) but a worse OS (HR = 1.09, 95%CI: 1.01–1.18, p = 0.040) compared to that of the non‐CLBC patient, but there was no significant difference both BCSS (HR = 1.05, 95%CI: 0.88–1.24, p = 0.614) and OS (HR = 1.08, 95%CI: 0.97–1.20, p = 0.168) after PSM.
01
问题一
这里有几个问题。
第一，倾向性评分匹配是怎么做的，方法部分没有任何描述。哪些指标纳入了评分，匹配的卡尺也就是差别多少就算配上了，作者没有任何描述，鬼晓得这个分析结果是否靠谱。
这里，和大家介绍一下研究规范：大家知道临床试验有CONSORT规范，观察行研究有 STROBE 规范，荟萃分析有 PRISMA 规范，倾向性评分匹配分析同样也有成文的、被全世界接受的规范，写什么/不写什么，在哪里写，都有明确的规定。
第二，PSM分析的基本逻辑是匹配那些可能影响治疗方案选择的因素，其目的和临床试验中的随机对照一样，既然作者做了匹配，2个组别的结局变量就是一个简单的对比，再用多因素回归就是脱裤子放屁，多此一举了。
问题二
03
第二句描述的是接受保乳手术的中央型乳腺癌和非中央型乳腺癌患者的结局，没有任何意义，谁高谁低，和这个研究的目的（中央型乳腺癌患者保不保乳）没有一毛钱关系，不应当出现在这篇论文，尤其是论文摘要这样重要的地方。占用的篇幅应当节省出来，描述我们刚才说到的问题，中央型乳腺癌患者多少例，保乳和不保乳手术的分别多少例，PSM是怎么做的。
Conclusion
Our findings revealed that BCT should be an acceptable and preferable alternative to mastectomy for well‐selected patients with CLBC.
01
问题一
“reveal”和“should”含义是不匹配甚至相互矛盾的，你不可以说“我的调查结果显示这事儿应当这么做”。
问题二
02
“well‐selected patients with CLBC”什么含义？没有人能够理解。“well-selected – 经过仔细挑选的”，也就是不是所有的中央型乳腺癌，具体是哪一部份，没有远处和腋窝淋巴结转移？病变大小有什么讲究没有？你是不是可以不要卖这个关子？
转自“白话医学研究”微信公众号，本文仅作为学术交流分享，如有侵权，请联系本站删除！
