Talk:Pituitary apoplexy/GA1

GA Review
The edit link for this section can be used to add comments to the review.''

Reviewer: Axl (talk · contribs) 18:51, 26 September 2011 (UTC)

Okay, let's do it. Axl ¤  [Talk]  18:51, 26 September 2011 (UTC)


 * Thanks very much! JFW &#124; T@lk  19:33, 26 September 2011 (UTC)

From the lead section, paragraph 1: "This usually occurs in the presence of a tumor of the pituitary, although in 80% of cases this may not have been diagnosed previously." Should this be "...in 80% of cases this will not have been diagnosed previously"? Axl ¤  [Talk]  18:52, 26 September 2011 (UTC)


 * Changed to "has not been diagnosed previously", which is more concise. JFW &#124; T@lk  19:33, 26 September 2011 (UTC)

From "Acute symptoms", paragraph 1: "The most common symptom, in 95–100% of cases, is a sudden-onset headache...". How about "The most common symptom, in over 95% of cases, is a sudden-onset headache..."? Axl ¤  [Talk]  18:55, 26 September 2011 (UTC)


 * Agree that a range that includes 100% can easily be replaced with "over ~" to sound less formal. JFW &#124; T@lk  19:33, 26 September 2011 (UTC)

From "Acute symptoms", paragraph 1: "This occurs in a median of 24% of cases." I'm not sure that the average reader would know what "median" means (no pun intended). Actually I'm not even sure why the term "median" is used here. The percentage is not countable. An individual either has meningism or doesn't, right? Why not just say "This occurs in 24% of cases"? Axl ¤  [Talk]  19:03, 26 September 2011 (UTC)


 * Agree. This was based on Nawar et al, who in their review made sure to calculate medians for most data from smaller case series. For the average reader it is irrelevant. JFW &#124; T@lk  19:33, 26 September 2011 (UTC)

In "Acute symptoms", I wonder about the diagram demonstrating bitemporal hemianopsia. Would the average reader understand what it means? The left and right visual fields are not labelled. Anyway, I don't think that this is important enough to be changed for GA status. Axl ¤  [Talk]  19:59, 26 September 2011 (UTC)


 * I was hoping that readers truly interested in the background would click on bitemporal hemianopsia and learn more about the intricacies of the anatomy of the optic nerve. JFW &#124; T@lk  20:58, 26 September 2011 (UTC)
 * Why have you marked this as "done"? I don't think that anything has been done. Axl  ¤  [Talk]  20:23, 27 September 2011 (UTC)


 * Sorry, must have done that semi-automatically. Do you think a long treatise on the anatomy of the optic chiasm would be beneficial to this article? JFW &#124; T@lk  21:20, 27 September 2011 (UTC)
 * There is no need to be sarcastic. The left and right visual fields should be labelled. So should the directions medial and lateral. And a link to "Visual field" would be helpful. I don't have the experience with image editing (Photoshop?) to do this. Axl  ¤  [Talk]  23:58, 28 September 2011 (UTC)


 * I did not intend to sound sarcastic at all! I was asking the honest question whether the text is currently lacking with regards to explanations for the patterns of visual field defects; unfortunately a lot of anatomy will need to be discussed if we want to go round that route. I will try to improve the image caption. JFW &#124; T@lk  19:42, 1 October 2011 (UTC)
 * The longer caption is only moderately helpful. The image itself needs to be labelled with "Left visual field" and "Right visual field", with the directions "medial" and "lateral" (or "nasal" and "temporal"). If this labelling isn't forthcoming, the article would be better off without the picture. Axl  ¤  [Talk]  21:11, 6 October 2011 (UTC)


 * I have uploaded a version of the image with text-based labels, as suggested. It's in one of my favourite fonts, Zapf Humanist. On thumbnail it is quite small. JFW &#124; T@lk  22:13, 6 October 2011 (UTC)
 * Perhaps add "Right visual field" and "Left visual field" above the visual fields? Axl  ¤  [Talk]  22:32, 6 October 2011 (UTC)


 * Will do that when I can access my image editor at home (probably tomorrow night). JFW &#124; T@lk  10:12, 7 October 2011 (UTC)


 * Sorted. JFW &#124; T@lk  20:05, 8 October 2011 (UTC)
 * Thanks. I have adjusted the caption. Axl  ¤  [Talk]  19:39, 10 October 2011 (UTC)

From "Causes", paragraph 1: "the risk was higher in ... tumors that were growing more rapidly (over 3.5 mm)." Is this 3.5 mm per year? Axl ¤  [Talk]  19:55, 26 September 2011 (UTC)


 * Strangely, the source doesn't say whether this is change per year or any change. I think we may need to drop the "3.5 mm" numeral because the source is so unclear about it! JFW &#124; T@lk  20:58, 26 September 2011 (UTC)
 * Okay, please remove it. Axl  ¤  [Talk]  21:32, 26 September 2011 (UTC)

From "Causes", paragraph 2: "radiation therapy to the pituitary [predisposes to apoplexy]." Pituitary radiotherapy is only given to patients with known pituitary adenomas, typically those who are unsuitable for surgery or who developed recurrence following surgery. Is pituitary RT really an independent risk factor, or is it just a marker of significant underlying pituitary disease? Axl ¤  [Talk]  20:08, 26 September 2011 (UTC)


 * Excellent point, but the source is not clear. I think cranial radiotherapy that (unintentionally or inevitably) involves the pituitary can also cause this. The lists of causes, although highly consistent between the major sources, are all based on case reports and small case series. It would be hard to disentangle! JFW &#124; T@lk  20:58, 26 September 2011 (UTC)
 * Well, if the source says that pituitary RT is a risk factor and we don't have any better evidence, we can leave it as it is. Axl  ¤  [Talk]  21:34, 26 September 2011 (UTC)

From "Mechanism", paragraph 1: "Larger tumors are more prone to bleeding, and more rapidly growing lesions (as evidenced by detection of increased levels of the protein PCNA) may also be at a higher risk of apoplexy." This was previously mentioned the "Causes" section. I don't think that it needs to be repeated here. The significance of PNCA should be mentioned elsewhere instead, perhaps in a "Prevention" section. Speaking of which, doesn't pituitary surgery reduce the risk of apoplexy? If so, this should be mentioned in a "Prevention" section. And what about treatment of hypertension? Axl ¤  [Talk]  20:19, 26 September 2011 (UTC)


 * I felt that the mention of PCNA was a bit off-topic for "causes" but fit right in the "mechanism" section. Let me know if you think that duplication must completely be avoided here.
 * None of the sources directly mention surgery as a way of preventing apoplexy. To place things in perspective, apoplexy is still a very rare event in absolute terms, and transsphenoidal hypophysectomy is not exactly completely blameless surgery.
 * Similarly, none of the sources say in so many words that good BP control prevents apoplexy, presumably because data are lacking. JFW &#124; T@lk  20:58, 26 September 2011 (UTC)
 * Okay, you're saying that there isn't enough information available to justify a "Prevention" section. Let's leave it as it is. Axl  ¤  [Talk]  21:36, 26 September 2011 (UTC)

From "Treatment", paragraph 2: "This is followed by the administration of hydrocortisone, which is pharmaceutical grade cortisol, by a four times daily injection into the muscle, intravenously by a continuous infusion, or intermittently." I don't think that the frequency of treatment or the various routes need to be mentioned. Axl ¤  [Talk]  20:46, 26 September 2011 (UTC)


 * You're right, I'll drop that. JFW &#124; T@lk  20:58, 26 September 2011 (UTC)
 * I've retained the route of administration. Oral hydrocortisone is not advised by any of the sources, and explaining the word "parenteral" would amount to explaining what IM and IV actually is! JFW &#124; T@lk  21:08, 26 September 2011 (UTC)
 * Okay, that's fine. Axl  ¤  [Talk]  21:40, 26 September 2011 (UTC)

I've been looking through the literature and I found this article. The study found no correlation with hypertension. I looked again at the Wikipedia article and the "Causes" section indicates that a quarter have a history of hypertension. However this is the same as the prevalence of hypertension in the background population. Axl ¤  [Talk]  21:31, 26 September 2011 (UTC)


 * I would have difficulty displacing three secondary sources on the basis of one primary source, but in reality it is probably correct. 25% is possibly an overestimate of the population prevalence of hypertension. JFW &#124; T@lk  22:12, 26 September 2011 (UTC)


 * Have you seen this article? It states "However, whereas diabetic ketoacidosis and malignant hypertension may precipitate an apoplectic episode, there is no evidence that diabetes or hypertension are more common in patients with pituitary apoplexy." Axl  ¤  [Talk]  07:28, 27 September 2011 (UTC)


 * Interesting... now we have several primary studies saying that hypertension is probably not a factor, yet all recent reviews include it, including the very recent British guideline document. Its pronouncements are based on studies from 1999 and 2004, ignoring the JNNP article completely. Clearly, the data are conflicting. Again I would not want to use primary studies to displace the conclusions of secondary sources, particularly if there are several. JFW &#124; T@lk  08:02, 27 September 2011 (UTC)
 * The Wikipedia article currently states "A quarter has a history of high blood pressure", referenced to Rajasekaran. Those UK guidelines state "Hypertension was by far the commonest (26%) predisposing factor." I'm certainly not disputing the finding that 26% of apoplexy patients had hypertension. However I am not convinced that this is any different from the background population, in which case it isn't really a predisposing factor at all. This article indicates a prevalence of 28.9%. Wikipedia's "Hypertension" article indicates an overall prevalence of 26%. I believe that Rajasekaran has mischaracterized hypertension's role as a risk factor. Axl  ¤  [Talk]  20:17, 27 September 2011 (UTC)


 * I can't disagree with your assessment, but at the moment, all we can do is blame it on the sources. It's not just Rajasekaran, but all other sources, that repeat the mantra about hypertension. Do you fancy submitting a letter to Clinical Endocrinology to rectify the matter? :-) JFW &#124; T@lk  21:20, 27 September 2011 (UTC)


 * Okay, I have Swearingen & Biller. Here is what is says: "Based on these theories of pathophysiology, a number of comorbid conditions, procedures, and drugs associated with the onset of pituitary apoplexy have been described as predisposing factors, though causality is difficult to establish. A list of such factors associated with pituitary apoplexy in the medical literature and accompanying references is found in Table 2." Table 2 is a long list, with hypertension at the top of the list.


 * The text continues: "Among the most commonly cited factors are those that might reasonably compromise the integrity of tumor vessels, impair effective hemostasis, or cause rapid changes in intravascular pressure. These include hypertension, altered coagulation or platelet function, radiation therapy of known pituitary tumors, head trauma, and surgery."


 * The end of the section states: "While some of the reported predisposing factors seem more plausible than others, no causative relationships have been proven. Clearly, many of the predisposing factors, such as diabetes mellitus and hypertension, are very common and a causative relationship would be nearly impossible to establish given the rarity of apoplexy."


 * I think that we should take a similarly cautious approach towards our "Causes" section. Axl  ¤  [Talk]  19:07, 30 September 2011 (UTC)


 * I think we can certainly quote this source in the right context. JFW &#124; T@lk  19:42, 1 October 2011 (UTC)


 * I see that you added this statement towards the end of the "Causes" section: "While these associations have been reported, it is uncertain whether they are in fact causative. High blood pressure, for instance, is a common problem in the general population, and may not significantly increase the risk of apoplexy." Actually I think that this needs to be at the start of the paragraph. Axl  ¤  [Talk]  17:50, 3 October 2011 (UTC)


 * I chose this formulation because all other reviews I used placed quite a lot of emphasis on the associations. As you have seen, some sources have an even larger list based almost entirely on single case reports. What Russell & Klahr Miller do is place these associations in a bigger framework and advise caution, rather than the other way round. JFW &#124; T@lk  20:09, 3 October 2011 (UTC)
 * I am particularly concerned that the title of the subsection is "Causes". The TL:DR crowd may read "high blood pressure" and "coronary artery bypass graft", without appreciating the context of the ambigiuity. How about putting "Many associations have been reported, but is uncertain whether they are in fact causative" as the second sentence of the paragraph? Axl  ¤  [Talk]  19:32, 5 October 2011 (UTC)


 * I have tried a rephrase. Hope this settles the issue. JFW &#124; T@lk  21:14, 5 October 2011 (UTC)
 * That's good. Thanks. Axl  ¤  [Talk]  21:32, 6 October 2011 (UTC)

From "Treatment", paragraph 3: "Surgery is most likely to improve vision if there was some remaining vision before surgery, and if surgery is undertaken within a week of the onset of symptoms." Should this be in the "Prognosis" section? Axl ¤  [Talk]  20:31, 27 September 2011 (UTC)


 * Borderline. I think that because it is conceptually related to the decision to operate, it is better discussed in context. JFW &#124; T@lk  21:20, 27 September 2011 (UTC)
 * Hmm, okay. Axl  ¤  [Talk]  17:57, 3 October 2011 (UTC)

"Prognosis", paragraph 1: "Before the introduction of steroid replacement, the mortality from pituitary apoplexy approximated 50%. This has now improved significantly." The historical mortality should be in the "History" section. "Improved significantly" isn't helpful. Surely there are some modern figures? Axl ¤  [Talk]  20:49, 27 September 2011 (UTC)
 * This study found that of 16 patients with apoplexy, three died. Axl  ¤  [Talk]  21:01, 27 September 2011 (UTC)
 * Better still, this book found a mortality of 1.6% (n=353). Axl  ¤  [Talk]  21:08, 27 September 2011 (UTC)


 * Good find! Will move the prognostic data to the "prognosis" section. JFW &#124; T@lk  21:47, 27 September 2011 (UTC)

In "Prognosis", what proportion develop GH deficiency? Axl ¤  [Talk]  21:32, 27 September 2011 (UTC)


 * Neither of the sources gives an exact figure, although Nawar says "practically all" without a quantifier. JFW &#124; T@lk  21:47, 27 September 2011 (UTC)
 * Okay. Swearingen/Biller doesn't quote a figure either. Axl  ¤  [Talk]  18:04, 3 October 2011 (UTC)

From "Prognosis": "Before the introduction of steroid replacement, the mortality from pituitary apoplexy approximated 50%." This should be in the "History" section. Also, it doesn't flow into the next sentence that states that the mortality in larger case series was 1.6%. Axl ¤  [Talk]  18:01, 3 October 2011 (UTC)


 * Agree. JFW &#124; T@lk  20:09, 3 October 2011 (UTC)

From "Epidemiology": "Pituitary apoplexy is considered rare." How about: "Pituitary apoplexy is rare." Axl  ¤  [Talk]  18:06, 3 October 2011 (UTC)


 * Agree. Unfortunately, none of the sources seemed to have a figure of annual incidence. JFW &#124; T@lk  20:09, 3 October 2011 (UTC)
 * This is from Swearingen/Biller:-


 * Surgical series of pituitary adenomas and apoplexy are biased to include patients with larger or more symptomatic tumors, as these are routinely referred for surgical consultation, whereas many prolactinomas and nonfunctioning microadenomas are not. It might therefore be expected that the rate of pituitary apoplexy would be lower in an unselected series of true incidentalomas. Surprisingly, this was not the case in a prospective study of patients with pituitary adenomas found incidentally by magnetic resonance imaging (MRI). Each patient underwent a complete endocrinologic evaluation and visual perimetry and was excluded from the analysis if there was evidence of endocrine hypersecretion, visual field defect, or panhypopituitarism. A total of 42 patients were included in the study, and their adenomas designated as "incidentally found", although 13 patients had the initial MRI performed due to chronic headache. During a mean follow-up period of 62 months, four patients (9.5%) developed classically symptomatic pituitary apoplexy with evidence of new hemorrhage within the tumor by MRI.


 * This is the reference. Axl  ¤  [Talk]  21:57, 6 October 2011 (UTC)


 * I have instead referenced a larger meta-analysis, which is presumably better. What I meant was the fact that no study has attempted to calculate the number of people per 100,000 of the population who develop pituitary apoplexy. This can be helpful in setting one's index of suspicion. JFW &#124; T@lk  22:18, 6 October 2011 (UTC)
 * You are right. Although from Swearingen/Biller:-
 * No population-based reports of the incidence of pituitary apoplexy have been published. However, there are data available on the prevalence of pituitary tumors, the substrate for pituitary apoplexy.... If we extrapolate an incidence rate of apoplexy in nonsurgically treated macroadenomas of 1.8% per year from the data of Arita et al. (9.5% incidence over 62 months) and assume a prevalence of macroadenomas of approximately 0.1%, the annual incidence of apoplexy is predicted to be approximately 18 per million. This is higher incidence than clinical experience would suggest is the case. A population-based study of apoplexy incidence would be necessary to resolve this apparent paradox. It seems likely that the rate of pituitary apoplexy will be lower in populations with better access to health care, and therefore fewer macroadenomas left undetected and untreated.
 * Axl ¤  [Talk]  22:41, 6 October 2011 (UTC)


 * Excellent. I will put that in. JFW &#124; T@lk  10:12, 7 October 2011 (UTC)
 * Thanks. Axl  ¤  [Talk]  19:41, 10 October 2011 (UTC)

From "Epidemiology": "Based on calculations from different case series, the mean age of people with the condition is 42.5 or 50.9." I don't think that it's helpful to quote two such precise and conflicting values. Swearingen/Biller states "50 years". Also, I think that "average" would be better than "mean". Axl ¤  [Talk]  20:03, 5 October 2011 (UTC)


 * Sorted. JFW &#124; T@lk  21:14, 5 October 2011 (UTC)
 * Thanks. Axl  ¤  [Talk]  21:43, 6 October 2011 (UTC)

Summary
Okay, I think that we're done.


 * 1) The article is well-written and fits with WP:MOSMED.
 * 2) The text is accurate, with good quality sources.
 * 3) The article covers the whole topic.
 * 4) The article is neutral.
 * 5) There is no edit war or controversy.
 * 6) The pictures are appropriately used. (I wonder if the labelled bitemporal hemianopsia picture could be uploaded to Wikimedia Commons?)

The article meets the GA criteria. My thanks and congratulations to JFW. Axl ¤  [Talk]  20:25, 10 October 2011 (UTC)


 * Thanks very much, and pleasure working with you on this. Your review has significantly improved the article. JFW &#124; T@lk  20:37, 10 October 2011 (UTC)