前medhelp专家Dr. Hunter Handsfield对窗口期的新看法

fegaa

前几天有幸和Dr.Handsfield在他的新论坛上请教了一些他对窗口期改动的看法并且得到了回复.  Dr.Handsfield 和Dr Hook是前几年一直在medhelp专家板块帮助恐友解答问题的医生，也是美国顶级的艾滋病专家。 后来因为medhelp关闭了专家板块，他们去成立的自己的论坛，现在是 https://www.askexpertsnow.com/ 英文好的朋友可以去向他们提问题。  

几年期他们在medhelp上一直坚定认为的4代4周排除，但是在两年前他们的观点改到了6周。作为恐艾患者以及同行业人员，我向Dr.HHH 咨询了他改变观点的原因。 以下是我的问题和他的恢复。

他两年期给出的更改意见:
New data became available, that's all. Our opinions and advice always are based on the best available science, and scientific data often are not static as new research is done. A published review of all the available HIV tests (https://www.ncbi.nlm.nih.gov/pubmed/29140890) included data that in 1-2% of cases, it takes 6 weeks for the antigen-antibody (4th generation) HIV blood tests to be conclusive, and CDC published an accompanying ediorial that endorsed that conclusion (https://www.ncbi.nlm.nih.gov/pubmed/29140891). There is no "official policy in the US", but the CDC recommendations usually are highly respected. At that point we revised our advice accordingly.

大概意思是他们对窗口期的建议随着新的数据的出现而改变，一个最近发表的文章被美国CDC采用，文章根据现有数据总结了用第四代检测，1%-2%的人需要6周才能排除。并且指出美国没有官方认证的窗口期，CDC只是一个给出了高度权威的建议。

以下是我们的对话：

我的问题：
I also read the review article that CDC cited  regarding the new window period (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5718364/).   However, the data the original author used came from some old seroconversion panels sold by biocompanies.  It seems like the 99 percentile of the window period at 45 days was estimated based on statistical modeling of those seroconversion panels instead of recently reported cases (https://academic.oup.com/cid/article/64/1/53/2194435).  

So, if my understanding is right, why do some experts and countries still recommend 28 days as the window period, since these seroconversion panels have been long existing for 20+ years and readily available on google.   It seems impossible  that they just realized they had been wrong for many years.

How well do you think this study correlates with the real world medical practice?

My concern is that the false positives occur quite often at our facility and they are usually resolved by repeating or confirming.  Although there's probably no documented false negative,  according to the current testing algorithm, the first negative result will be the final result.   I know this kind of event doesn't happen as often as false positive , but if a false negative trully happened, it will be mistakenly reported as a true negative.

我读了那片论文，6周的窗口期引用的是生物公司贩卖的20年前的血清转阳数据，并且是通过统计模型计算出来的，而不是近期新发现的病例。既然这些数据存在了20多年，为什么近期才把窗口期从4周调到6周，还有为什么很多专家依然支持4代4周排除，我不是很理解为什么这些专家会无视这些存在了这么久的病例。 第二个问题是您觉着这项统计与现实的临床操作相关度有多高。 第三个问题是关于假阴，我们医院经常出现假阳，通常重复实验或者做下一步确诊就可以排除假阳的情况，但是基于现在的检测标准，如果第一次测是阴性，那么结果会被直接发出去，不会再做下一步调查。我不能确定阴性结果里面那些是真阴，那些是假阴。虽然我不认为假阴会像假阳那么普遍。

以下是他的恢复

The "new window period" isn't really new; no change in our advice for about two years. In any case, the "old" 28 day window still detects 98-99% of infections.  The first of the two links you provide is the resource on which our advice on this forum is based. Why do some sources stick with 28 days? They are not wrong to do so. Of all HIV testing done, a small minority is in situations following a particular exposure. Most testing is done in people at general risk, without regard to a particular exposure event. In that context, a test with 98-99% sensitivity at 4 weeks, and virtually 100% at 6+ weeks, is entirely acceptable.  There are several reasons that false negative results don't occur, at least not with the combinations of tests. As a scientist, I rely on data and am prepared to alter this perspective. But for now, while I understand the theoretical concern, I really don't believe it's a likely issue -- anywhere, any time.---

大概意思是他们在过去两年一直建议4代6周窗口期，但是4周已经可以检测出99%了。 那些认为4周排除的专家也不算是错，只是通常来讲hiv检测不是针对某一次的高危行为，而是更像普通常规查体。98%-99%的病例在4周就查出来了，6周就基本100%了。他用了Virtually 100%这个词。

我自己对这句话的理解是基于宏观公共卫生考虑，98%-99%就已经足够了，已经足够防止病毒扩散了，当然测到6周是几乎不会漏检的，要是在4周真有漏检的那么么一两个对社会危害也大不到哪去，也是可以接受的。不知道理解的对不对

他没回答我第二各问题

第三个关于假阴的问题，他认为不会发生，但是他的观点会随着新的数据的出现而改变，但是起码现在他对假阴的假设是否定的