Nothing Special   »   [go: up one dir, main page]
Wikipedia

Wikipedia:Peer review/Deep vein thrombosis/archive2

< Wikipedia:Peer review

Deep vein thrombosis

edit
Previous peer review
Toolbox
Article (edit | visual edit | history) · Article talk (edit | history) · Watch • Watch peer review

This peer review discussion has been closed.
I've listed this article for peer review because it passed GA and I want it to meet FA standards.

Thanks, Biosthmors (talk) 19:46, 9 July 2012 (UTC)[reply]

Comments

  • I'm not a medical expert, but I'll take a stab. The article's in good condition (har, har), but here's what I'm noticing.
  • Is there any compelling reason we need to abbreviate PE, VTE, PTS, GCS and ITE in the lead? That's a lot of abbreviation going on in a section that ideally should be as accessible as possible. I count only two uses of PE in the lead, no uses of VTE, two uses of PTS, one of GCS and none of ITE. I would suggest spelling these all out in the lead; they can be abbreviated as necessary in the body, which doesn't need to be as accessible.
  • Also, in the same sentence, is there any reason we say "has been applied" instead of "applies"? Is this something that has happened in the past but no longer does, or else only happened occasionally in the past? Could it be "is sometimes applied" if that's the case?
    • From memory, the source mentioned two different studies as using these cut-offs. I'm not sure how representative they are, so "applies" might be a little bit ORy. I'll see if I can find a clinical practice guideline cut-off. Biosthmors (talk) 22:06, 17 July 2012 (UTC)[reply]
  • In the image caption: "The popliteal vein is at the top of the image" could this be marked more clearly? The "top of the image" is rather inspecific, I would suggest.
  • What does the following mean? "but most of those with suspected DVT do not have it after evaluation" Are we saying "are determined not to have it after evaluation"? Is the point that it's often suspected but rarely present?
  • There's a lot of medical jargon in the Causes section. You might consider either including plain-English descriptions where they seem useful or else replacing medical terminology with more commonly understood wordings where possible. E.g. "deficiencies in the anticoagulation factors protein C, protein S, antithrombin, or mutations in the factor V and prothrombin genes". What do these things regulate/cause, in plain English?
    • Added some plainer expanatory text.
  • Same jargon issue with the following (is there an easy translation?): "the majority of venous thrombi form without any injured endothelium"
    • Reworded, thanks.
  • It would be nice if the remainder of the paragraph were made more accessible and less jargon-y. This isn't necessarily a requirement, but if it's doable without adding a ton of new text, I don't see why it shouldn't be done.
    • Made some changes.
  • The same goes for the rest of the section. I doubt if anyone lacking a medical background will be able to understand this completely.
    • I've reworded some things to make it more accessible. Maybe I can find other spots where the wording around the molecular biology info could be improved.

Comments - I'll add more later but in the meantime may I suggest that you replace the diagram of the coagulation cascade with this one File:Coagulation in vivo.png, which gives a better representation of what actually happens in vivo as opposed to in the laboratory. I have to own up to drawing this one, so there might be a perceived conflict of interest. I'll take a closer look at the Causes and D-Dimer sections later and add any, hopefully, constructive comments here. My first impression is that these sections are comprehensive, but the bulleted lists might work better as tables. I haven't looked yet, but I would ensure that there are no primary studies cited and that all sources are compliant with WP:MEDRS. Graham. Graham Colm (talk) 21:43, 24 July 2012 (UTC)[reply]

Thank you for your comments. The overwhelming majority of sources are either marked as reviews in databases or are clinical practice guidelines. There might be a quibble over using doi:10.1177/1538574411432145, for example, in the text as I do but I tried to cite it as a secondary source despite it not being marked as a review in a database. It supports "In upper extremity DVT, central venous catheters are a dominant risk factor". I'll double check that. (now sourced to a review) As for the images (thank you for creating one) they both appear to contradict the text a bit by being biased towards arterial thromboi by either making platelets or endothelial damage (instead of endothelial activation {potential COI -- I created that one}, but maybe some authors would consider that "damage") a necessary precursor to tissue factor production. I could change the image but I think the emphasis on platelets may contradict the article text more than the other image. Any other opinions on the appropriateness of the images as they relate to venous thrombosis? Biosthmors (talk) 19:25, 26 July 2012 (UTC) Biosthmors (talk) 20:54, 10 August 2012 (UTC)[reply]
A problem with the current image is that it illustrates the artificial cascade in vitro and not what is thought to occur in vivo. Do activated platelets not play a role in DVT? (Ruggeri ZM (2007). "The role of von Willebrand factor in thrombus formation". Thrombosis Research. 120 Suppl 1: S5–9. doi:10.1016/j.thromres.2007.03.011. PMC 2702526. PMID 17493665.) Graham Colm (talk) 20:13, 26 July 2012 (UTC)[reply]
(ec) Great question. Research suggests they could (in humans perhaps).[1][2] I went ahead and changed the picture. If I find a source that says platelets don't play a role (which I doubt could be said) I'll revisit this. Biosthmors (talk) 20:27, 26 July 2012 (UTC)[reply]
If there is a way to incorporate protein C, S and antithrombin into the image you created I think that would be a plus for this article at least. Biosthmors (talk) 20:31, 26 July 2012 (UTC)[reply]
Yes, this has been on my to do list for a while (with regards to Coagulation). I need to find a way that doesn't cause too much clutter. I'll come up with something at the weekend. Graham Colm (talk) 20:42, 26 July 2012 (UTC)[reply]
That sounds great, thanks. Biosthmors (talk) 20:50, 26 July 2012 (UTC)[reply]
I have added the modulators antithrombin and activated protein C (in green) to the diagram above the factors they regulate. Graham Colm (talk) 12:59, 29 July 2012 (UTC)[reply]
Excellent, thank you. Biosthmors (talk) 15:47, 29 July 2012 (UTC)[reply]

I am surprised to see that the lead section states "Other risk factors include older age (the strongest)". I have been trying to find a source that states relative risk/odds ratios for the risk factors. Here is one. The paper quotes "advancing age" as a weak risk factor. (Curiously, cancer is not quoted as a risk factor). Axl ¤ [Talk] 11:21, 26 July 2012 (UTC)[reply]

Hmm, this reference implies that aging is a major risk factor. Axl ¤ [Talk] 11:35, 26 July 2012 (UTC)[reply]
Okay, after looking through a few more references, I concede that age is an important risk factor. However, which reference indicates that age is "the strongest"? Axl ¤ [Talk] 18:49, 26 July 2012 (UTC)[reply]
Good catch. A useful chart can be found in Lijfering (with relative risks). It (oddly to me) gave the RR of aging as 1 to ∞. I made edits to make sure the lead didn't say anything the article did not. Biosthmors (talk) 19:25, 26 July 2012 (UTC)[reply]
The epidemiology section also emphasizes this aspect. Biosthmors (talk) 20:09, 26 July 2012 (UTC)[reply]
Lijfering does not say that age is the strongest risk factor. Indeed it includes the caveat: "Another simplification is that 'increasing age' is used as a container concept here, that is a mix of unknown and known risk factors that either become stronger with age or become more prevalent with age." The table in Lijfering quotes a relative risk range of "1 to ∞" for age, compared to the general population. Lijfering refers to this article by Naess. Unfortunately I don't have full access to it. Naess describes incidence. An infinitely high increased risk of incidence would mean that the at-risk population (the oldest group) would all be continuously developing new DVTs. Part of the confusion about age as a risk factor is that it, unlike every other risk factor in Lijfering (with the possible exception of air pollution), is a continuous variable. Relative risk should only be applied to a discrete age group. Axl ¤ [Talk] 21:32, 26 July 2012 (UTC)[reply]
True. I wrote this thinking I was just summarizing the ∞ sign (though it is illogical as you explain) and the epidemiology section which states incidence goes up by a factor of ~1000 during a normal life. A thousand is definitely higher than any other risk factor so I thought it was a safe summary. Biosthmors (talk) 22:00, 26 July 2012 (UTC)[reply]
I downloaded a copy of Naess. I could email it to you. Biosthmors (talk) 23:07, 26 July 2012 (UTC)[reply]
Yes please! Axl ¤ [Talk] 23:12, 26 July 2012 (UTC)[reply]
Thank you for sending me the journal paper. Naess does not say that any population group has an infinitely high risk. It says "The incidence rates increased exponentially with age. Incidence rates in subjects aged 70 years or above were more than three times higher than those in subjects aged 45 to 69 years, which again were three times higher than the rates in subjects aged 20–44 years." Even if we take the statement "incidence rates increased exponentially with age" at face value, the "infinitely high risk" than Lijfering quotes would only be reached at infinitely old age. Axl ¤ [Talk] 19:14, 27 July 2012 (UTC)[reply]
Agreed. Page 533 of Bovill (cited in Causes section) states "Aging is the strongest risk factor for VT: Thrombosis risk rises exponentially from an incidence of 1/10,000 in young individuals to an incidence of 9/1,000 by 80 years of age (44)." Reference 44 is doi:10.1182/blood-2007-06-096545. Biosthmors (talk) 22:05, 30 July 2012 (UTC)[reply]
Are you sure that's the right link? Axl ¤ [Talk] 22:35, 30 July 2012 (UTC)[reply]
Yes, and that source (the one Bovill cites) states: "Incidence rates of VTE (Figure 1A) increase dramatically at about age 55 and by age 80 are nearly 1 in 100 per year, approximately 1000-fold higher than for those aged 45 or younger." Biosthmors (talk) 22:49, 30 July 2012 (UTC)[reply]
Ah, I thought that the link was supposed to be Bovill. I understand now. :-) Axl ¤ [Talk] 09:36, 31 July 2012 (UTC)[reply]
The statement from Silverstein refers to a combination of DVT and PE. Figure 1A shows a steep ("exponential") rise in the combined risk with age. However the line for DVT alone is much more shallow. This is implied by Silverstein's next sentence: "Furthermore, rates of PE rise faster than DVT in the elderly (Figure 1B) so that the disease has greater fatal impact." [Actually there is an error in the statement; it should refer to Figure 1A, not 1B.] Axl ¤ [Talk] 11:01, 31 July 2012 (UTC)[reply]
True, but I don't see that as contradicting the idea that aging is the strongest risk factor. For the graph, Silverstein 2007 cites Silverstein 1998. Table 1 of Silverstein 1998 shows an incidence of about 1 per 100,000 for those of age 0 to 14 and around 300 per 100,000 for the elderly (though using DVT only figures would understate the risk). No other risk factor comes close to increasing risk by 300X. Lijfering cites surgery, trauma, and immobilization as only increasing the risk of venous thrombosis by 5 to 50, the next largest value given in Table 1 there. Biosthmors (talk) 22:03, 31 July 2012 (UTC)[reply]
You continue to quote values for combined VTE to justify your claim about DVT. The is inappropriate, especially when both of the papers that you refer to (Silverstein 2007 & Silverstein 1998) show much less impact when considering DVT alone. I don't have access to Silverstein's 1998 paper in full, but the abstract states "The incidence of venous thromboembolism rose markedly with increasing age for both sexes, with pulmonary embolism accounting for most of the increase... the incidence of deep vein thrombosis remained constant for males across all age strata, decreased for females younger than 55 years, and increased for women older than 60 years."
I had a closer look at Lijfering's paper. It uses the phrase "venous thrombosis". It isn't obvious whether this refers specifically to DVT, or if it is inclusive of VTE. I looked at some of the references from Lijfering's table. Simioni, Blom and Grainge all describe combined VTE. Also, Lijfering has a long discussion about PE after air travel. I believe that Lijfering's paper is actually about VTE, not DVT. Axl ¤ [Talk] 22:52, 31 July 2012 (UTC)[reply]
I can email you Silverstein 1998. Lijering's table may well be for VTE, but the only way I see that mattering (for the statement that aging is the strongest risk factor for DVT) is if another risk factor simultaneously increases DVT risk while reducing PE risk. That seems impossible. But, because I haven't seen a source that specifically states aging is the strongest risk factor for DVT (just "venous thrombosis"), I don't have a problem removing it until a source says DVT. Biosthmors (talk) 23:15, 31 July 2012 (UTC)[reply]
I have already demonstrated that Lijfering's quoted risk for aging ("1 to ∞") is flawed. Even if you find a source that shows that aging is the "strongest" risk factor for VTE, extrapolation of that statement to DVT alone would be original research. Indeed the fact that Lijfering is about VTE, not DVT, makes its use as a reference for this article highly questionable. As a work-around, it may be possible to change some of the statements to be explicitly about VTE. For example, the "Risk factors" subsection could be changed to "Risk factors for venous thromboembolism". Axl ¤ [Talk] 10:15, 1 August 2012 (UTC)[reply]
Thanks for raising this issue. I will address it after I address some other comments first. Thanks. Biosthmors (talk) 17:44, 1 August 2012 (UTC)[reply]
Right now I've changed the title of the section to mention VTE but I've also copy-pasted the section at venous thrombosis so I'm not sure the entire risk factor list is needed at this article too. I will contemplate this further. Biosthmors (talk) 22:18, 3 August 2012 (UTC)[reply]
doi:10.1182/blood-2012-02-407551 is good on this issue. Biosthmors (talk) 20:56, 10 August 2012 (UTC)[reply]

From the lead section, paragraph 2: "Women have an increased risk during pregnancy (due to altered blood protein levels) and in the postnatal period, partially due to substances released by the placenta. However, some of those who develop DVT have no recognized risk factors." I don't understand why pregnancy is afforded so much detail when the preceding sentence contains a list of one-word risk factors. Axl ¤ [Talk] 21:40, 26 July 2012 (UTC)[reply]

Changed.

From the lead section, paragraph 3: "If anticoagulation is not possible, management may involve inferior vena cava filter placement to presumably prevent pulmonary embolism." IVC filter placement is rare. I recommend deleting this sentence from the lead section. Axl ¤ [Talk] 21:48, 26 July 2012 (UTC)[reply]

Removed. Biosthmors (talk) 21:55, 26 July 2012 (UTC)[reply]

From "Classification", paragraph 1: "An incident DVT is an initial episode." Is that really true? Black's Medical Dictionary (41st ed.) doesn't have an entry for this. Stedman's Medical Dictionary defines "incident" as "going towards; impinging upon, as incident rays". Axl ¤ [Talk] 22:04, 26 July 2012 (UTC)[reply]

That's what I'm seeing. Here's another example:[3] Biosthmors (talk) 22:17, 26 July 2012 (UTC)[reply]
And "incident (initial)" makes it more clear. Biosthmors (talk) 16:28, 27 August 2012 (UTC)[reply]

From "Classification", paragraph 2: "Pain and swelling are symptoms of acute DVT." I'm not sure why this is in "Classification" rather than "Signs and symptoms". Perhaps it should read: "Acute DVT is characterized by pain and swelling"? Axl ¤ [Talk] 22:25, 26 July 2012 (UTC)[reply]

Good idea. Done. Biosthmors (talk) 22:33, 26 July 2012 (UTC)[reply]

From "Classification", paragraph 2: "nonocclusive DVTs are more asymptomatic." Perhaps "nonocclusive DVTs are less symptomatic." Axl ¤ [Talk] 23:05, 26 July 2012 (UTC)[reply]

Done. Biosthmors (talk) 23:10, 26 July 2012 (UTC)[reply]

From "Signs and symptoms", paragraph 1: "Approximately half of people with DVT have symptoms. They include pain and tenderness in the leg." It isn't clear that "They" refers to the symptoms rather than the people until we read further. How about "These" instead of "They"? Axl ¤ [Talk] 19:25, 27 July 2012 (UTC)[reply]

Changed.

From "Signs and symptoms", paragraph 1: "They include pain and tenderness in the leg, swelling, warmth in the leg that is swollen or painful, redness or discoloration, and dilation of surface veins." How about "They include pain and tenderness in the leg, swelling, warmth in the affected leg, redness or discoloration, and distention of surface veins." Axl ¤ [Talk] 20:29, 27 July 2012 (UTC)[reply]

Changed.

From "Signs and symptoms", paragraph 1: "Most symptomatic individuals have another condition, and those possibilities include "cellulitis, Baker's cyst, musculoskeletal injury, [and] lymphedema"." I understand your concern about possible plagiarism, but I don't that is really applicable for a list of four items. The text would be better without the quotation marks and the bracketed conjunction. Axl ¤ [Talk] 18:08, 29 July 2012 (UTC)[reply]

Done, thanks.

From "Causes", paragraph 1: "Venous thrombi are recognized to be caused mainly by a combination of venous stasis and hypercoagulability." How about "Venous thrombi are caused mainly by a combination of venous stasis and hypercoagulability." Axl ¤ [Talk] 18:17, 29 July 2012 (UTC)[reply]

Done. Biosthmors (talk) 22:15, 30 July 2012 (UTC)[reply]

From "Causes", paragraph 2: "Acquired risk factors include the strongest risk factor—older age—due to changes in blood composition which favor clotting." Besides my concern about the first part of the sentence, the latter part implies that blood composition changes are the only cause of acquired risk factors. Venous stasis is actually an important feature of several acquired factors such as lower limb surgery, immobilization, pregnancy, abdominal cancer, etc. Axl ¤ [Talk] 11:51, 31 July 2012 (UTC)[reply]

Changed to "which alters blood composition to favor clotting". Biosthmors (talk) 22:21, 31 July 2012 (UTC)[reply]

From "Causes", paragraph 3: "Genetic causes of increased VTE risk include... non-O blood type." How important is non-O blood type? (I don't have access to the reference.) Would it be more accurate to say that O blood type is protective? Axl ¤ [Talk] 21:44, 10 August 2012 (UTC)[reply]

I will look up prevalence figures (I think they should go in the epidemiology section) and see if the source addresses this specifically. I think both statements are of equal accuracy logically, but all the sources I've seen treat it as a risk factor. FYI, feel free to request any source by email. I do not mind being asked to do so. Biosthmors (talk) 18:46, 11 August 2012 (UTC)[reply]
I cited incidence figures in the Epi section and I reworded the Causes section to reflect its importance. The wording for text in the Epi section is general while the data is U.S. centric, but I think it is probably OK that way. Biosthmors (talk) 21:12, 15 August 2012 (UTC)[reply]

From "Causes", paragraph 3: "Factor V Leiden, which prevents the Leiden variants of factor V from being inactivated by activated protein C." This doesn't make sense. The Leiden variant of Factor V is resistant to inactivation by APC. Axl ¤ [Talk] 22:05, 10 August 2012 (UTC)[reply]

Rephrased and sourced a freely accessible review. Biosthmors (talk) 21:14, 14 August 2012 (UTC)[reply]

Reference 31 (Lee) is a primary source. Axl ¤ [Talk] 22:15, 10 August 2012 (UTC)[reply]

Oops. Now replaced with a review. Biosthmors (talk) 19:34, 13 August 2012 (UTC)[reply]

From "Pathophysiology", paragraph 1: "There is a strong tendency for DVT to develop in the left leg (about 70 to 90% of the time)." Is this really true? (I don't have immediate access to the New England Journal reference at the moment.) I haven't seen or heard of this before. Axl ¤ [Talk] 22:42, 10 August 2012 (UTC)[reply]

I was surprised to read this the first time, and I am still surprised by the statement considering I've only seen it in this one source. I'll report what I find. Biosthmors (talk) 18:46, 11 August 2012 (UTC)[reply]
Marik cited a 1992 study about pregnancy, so I don't think the statement applies broadly. Removed. Good catch. Biosthmors (talk) 19:03, 13 August 2012 (UTC)[reply]

More comments

  • I suggest you check the "which" and "that" usage, which is inconsistent. "Which" without a preceding comma or bracket, should be "that".
  • Same with "may", which should be "might".
  • I think "probability assessment" needs explaining a little more; it sounds like jargon.
  • Done.
  • I think "dominated" is not the best word here "Treatment for DVT is dominated by anticoagulation ". Perhaps this could be written more clearly such as "mainly by"?
  • Done.
  • I didn't like this "At least one of these five thrombophilias". Does this mean "At least one of these five causes of thrombophilia."?
  • What is "pospartum hemoconcentration"? Does it mean "bleeding".
  • I would check the singular and plural usage of DVT. This sounds wrong, "There is a strong tendency for DVT to develop in the left leg."
  • WRT D-Dimers, there isn't really a positive test, but raised levels. Also these assays vary in sensitivity and specificity and most labs will only use one method. D-Dimer assays have limited value in hospitalised patients owing to co-morbidity.
  • Contrast agents pose a risk too, such as renal impairment.
  • One thing the source says is that it is "contraindicated in patients with renal insufficiency and severe allergic reactions to contrast medium". It also talks about getting into the dorsal foot vein, imaging ambiguity, dizziness and nausea. I just don't think we can list them all and the source I'm using at the moment doesn't seem to mention renal impairment exactly. Should I mention the renal contraindication? Biosthmors (talk) 20:46, 28 August 2012 (UTC)[reply]
I'm not sure, this is outside the scope of my practice. Is this reference useful? Nicholas BA, Vricella GJ, Smith M, Passalacqua M, Gulani V, Ponsky LE (February 2012). "Contrast-induced nephropathy and nephrogenic systemic fibrosis: minimizing the risk". The Canadian Journal of Urology. 19 (1): 6074–80. PMID 22316507. Graham Colm (talk) 21:11, 28 August 2012 (UTC)[reply]
Thanks. I checked but I don't have access to it. I doubt any more detail is needed in the sentence, as it seems best to only summarize that it is a benchmark but not very useful in practice. Biosthmors (talk) 22:44, 28 August 2012 (UTC)[reply]
  • In the Prevention section, I think it should say at the beginning that this depends on the level of risk.
  • I see later on we have "positive D-Dimer level", which is better. What is that level, ( I use >500 ng/ml).
  • On page e404S of Bates, in a note for a table about DD, in regards to the highly sensitive DD, it says, "Based on a specificity of 10.3% (95% CI, 6.6%-15.5%) and sensitivity of 100% (95% CI, 74.7%-100%) for the VIDAS DD using the standard cut point of 0.5 m g FEU/mL." I assume this is the same thing and google indicates FEU means fibrinogen equivalent units. Maybe another source will be more helpful for describing the basic clinical chemistry of DD. Biosthmors (talk) 21:29, 16 August 2012 (UTC)[reply]
This source might be useful [5]. It is possible that some readers will look up the article for an interpretation of a D-dimer result (number), and will be disappointed. I think we need to address this difficulty. Graham Colm (talk) 19:04, 18 August 2012 (UTC)[reply]
I've cited that article and Mayo source that has the levels listed. I plan on adding a note to specify what a positive level is. Biosthmors (talk) 21:08, 28 August 2012 (UTC)[reply]

This is an excellent article by the way. I might have more comments to add later. Graham Colm (talk) 11:36, 27 July 2012 (UTC)[reply]

  • periods in the abbreviated journal names (Ann Intern Med or J. Thromb. Haemost.?); consider using full journal names in this general-purpose encyclopedia with no space contraints (the lay reader may not know BMJ, CMAJ, or JAMA)
  • Periods used.
  • author name formatting: compare "John T. Owings" vs. "Nigel Key; Michael Makris; Denise O'Shaughnessy; David Lillicrap" (no et al. after 3 authors like most others) vs. "Esther S.H. Kim and John R. Bartholomew."
  • there's a mixture of title case and sentence case used with journal article titles
  • is the page range format "192–3" (ref #95), "194–95" (#96), or "180–189" (#98)
  • The section called "Notes" only has a few notes, but has many references. Consider pulling out the notes into a separate section, with the other section called "References" containing subsection "Cited literature"
  • Comments

In general: very good article, no doubts you will be able to take it to FA level

  • Double quotes are used quite often, in most cases presumably to indicate the sentence is a direct quote from the article being referenced. For improved readability I would prefer getting rid of most of them, either by paraphrasing the article in question, or just leaving them out when just a few words are quoted.
  • I am not sure about the appropriateness of the evidence strength grades that are included throughout the article. Although this format would work great in a review article, for an encyclopedic article it might be just a bit too much? Besides, this is not done in any other wikipedia article that I know of and it relies very heavily on one source. However it would be good to hear the opinion of others about this.
  • I understand the concern. I tend to like them as abbreviations to say "weak recommendation on decent evidence" or "weak recommendation on not so great evidence". I am also interested to hear other thoughts. I did add this to give the grades more context and meaning, and I will see if there are opportunities to remove specific recommendations and grades without removing information. Biosthmors (talk) 19:34, 16 August 2012 (UTC)[reply]
  • The diagnosis section could use an introduction explaining the difficulty in diagnosing DVT and how the wells score, d-dimer and imaging are used in combination to come to a diagnosis.

--WS (talk) 14:42, 31 July 2012 (UTC)[reply]

Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Peer_review/Deep_vein_thrombosis/archive2&oldid=1075158854"