【转帖】一篇专业顶级杂志letter to editor的诞生小记
guodanni 达人已点赞这个帖子发布于 13 年零 302 天前，其中的信息可能已发生改变或有所发展。
前几天收到了本专业顶级杂志Radiology的接收信，一个月前投的一篇letter to editor终于成功被接收了。虽然不是什么大文章，仅300多字的letter，但是这也是像我这种小医生能够第一次在本专业最顶级的杂志上发表观点，实在是小小的激动了2分钟。
其实回顾这篇小文章的诞生经历，颇觉幸运。话说那是一个月前的周五晚上，LP去外地出差了，呆在家里实在没事，正好看到8月份最新的radiology杂志刚刚刊出，于是随意浏览了一下，发现一篇很有意思的文章《Dry pleural dissemination in non-small cell lung cancer: prognostic and diagnostic implications》，介绍了干性胸膜转移（无胸腔积液）与湿性胸膜转移（合并胸腔积液）的生存情况，其中影像学对于判断胸膜转移有较大意义。这篇文章的通讯作者是来自韩国的Kim教授，也算是本专业的大牛之一了。这篇文章整体来说写的很不错，尤其是统计学处理，一看就知道很专业。
正在拜读之际，突然发现该文的一个明显错误。其实对于胸部成像而言，窗宽窗位的设置是很重要的，作者提出肺窗的中心是2700HU，怎么可能，最高也不过2000HU呀，这种小错误出现在顶级杂志实在是不应该，是不是有必要写封letter呢？当然，仅仅这点笔误是不够发表的，于是又仔细拜读了一遍，发现两组生存率比较过程中，接受到的临床治疗明显不同，P<0.05，总所周知，现在靶向性治疗满天飞，介入治疗也方兴未艾，系统治疗虽然传统，但是正受到越来越多的质疑。我并不是怀疑Kim教授的结论，但是我认为临床治疗的不同有可能会对预后产生影响，而不是单纯的用选择性偏移来解释。
另外，CT对于胸膜转移的诊断，越来越多的专家认为应该选择薄层高分辨的成像方式，这也是该篇回顾性分析文章所不具备的。
于是，趁着还有几分思路在，提笔写下了一封letter给radiology的编辑，全文如下。
High-resolution CT as an optimal diagnostic tool for evaluation of dry pleural dissemination in non-small cell lung cancer
Editor:
We read with great interest the article by Dr Kim and colleagues (1), “Dry Pleural Dissemination in Non-Small Cell Lung Cancer: Prognostic and Diagnostic Implications,” which appeared in the August 2011 issue of Radiology.
In that retrospective study, the median survival after initial presentation in patients with dry pleural dissemination (DPD) was almost two years longer than that in patients with wet pleural dissemination (WPD). Although nearly 90% primary lung cancers were adenocarcinoma in both groups, the DPD group received different treatment after diagnosis from the WPD group (P < 0.05) (1). This might have influenced the prognosis besides selection bias.
In spite of the possibly overestimated CT sensitivity mentioned in the limitation section, we agree with the conclusions of Dr Kim and colleagues that CT helps preoperatively clarify the presence of DPD. Based on our experience and the results of previous reports dealing with DPD, we think that high-resolution CT (HRCT) is an optimal diagnostic tool for the evaluation of DPD in lung cancer. Narrow slice width (usually 1~2 mm) and high spatial resolution image reconstruction algorithm (2) yield its high sensitivity and specificity for detecting subtle pleural dissemination. Actually, the incidence of pleural dissemination was relatively high in patients with pulmonary adenocarcinoma in whom the primary lesion was seen to be contiguous with the pleural surface (3). However, sometimes it is hard to detect DPD because of the partial volume effect and the growth way of pleural nodules within fibrous tissues. The most important advantage of HRCT is that a correct diagnosis at the initial presentation can prevent futile surgery (4). Special attention should be paid not only to previously reported CT findings of DPD, including fissural or pleural nodules and uneven bandlike pleural or fissural thickening but to the irregular and uneven pleural surface with convexities and concavities, and clusters of 2~3 mm nodules (5).
Additionally, the lung window (window width 1500 HU; window level 2700 HU) reported in Image Analysis seems erroneous.
接收后，收到了Yi Kyung Kim, MD的回信，全文如下。
RE: High-resolution CT as an optimal diagnostic tool for evaluation of dry pleural dissemination in non-small cell lung cancer
We thank Dr. Jiang and colleagues for their interest in our article (1). They noted that “the dry pleural dissemination (DPD) group received different treatment after diagnosis from the wet pleural dissemination (WPD) group, and expressed concern that this might have influenced the prognosis and created a selection bias.” We agree to their comments. The main cause of the different treatment the two groups received is that some of the patients in the WPD group could not endure chemotherapy due to their poor general condition. In fact, some patients in the WPD group died within a few weeks after the initial presentation, even before starting the chemotherapy. Therefore, even if the patients had been treated with chemotherapy, the treatment might not have significantly influenced on the prognosis in WPD patients. Furthermore, the adjusted hazard ratio for DPD to WPD was 0.283 (95% CI: 0.124, 0.647) (P= .003) when adjustments were made for covariates includingtreatment after diagnosis (1). This result indicates that DPD as compared with WPD is independently associated with a 72% reduced hazard rate for death after the initial presentation.
The diagnosis of DPD can be made when a subpleural or subfissural lung nodule or mass (primary lung adenocarcinoma) is associated with fissural or plural nodules, uneven band-like pleural or fissural thickening, irregular and uneven pleural surfaces, or clusters of 2-3-mm-sized pleural nodules at CT (1-3). One caveat is that some benign conditions, such as intrapulmonary lymph nodes, anthracotic or anthracofibrotic nodules and granulomas, may appear as small pleural or fissural nodules and thus simulate the imaging findings of DPD (2, 4, 5). Thus, obtaining a detailed past medical history which specifically notes prior pleural infection and occupational exposure to dusts or asbestosis is mandatory in order to avoid making a misdiagnosis of the DPD.
High-quality thin-section CT is a prerequisite for making the diagnosis of DPD. Appropriate scanning parameters should be selected for producing adequate CT images. In our institution, helical CT technique is used and the image data are reconstructed with high-frequency (bone window) algorithm for lung images and standard algorithm for mediastinal images, and with a section thickness of 2.5 mm or less. For viewing lung window images, we have window width 1500 HU and window level -700 HU. In our paper (1), the window level was written as 2700 HU which was a typographical error.
呵呵，人家真不愧是专家，解释的很到位，厉害。
(转自http://blog.sciencenet.cn/home.phpmod=space&uid=403432&do=blog&id=481012)
