We have discovered several errata for a New York City urologist, including in one paper that previously inspired one of our favorite headlines.
The latest development is pretty straightforward: Ashutosh K. Tewari has issued errata to multiple papers in two journals that note changes to some data points. But the backstory has some twists and turns, so you may need to read this one carefully.
We’ll start with the paper that might be familiar to eagle-eyed readers, on incontinence after surgery: “Effect of a Risk- Grade of Nerve-sparing Technique on Early Return of Continence After Robot-assisted Laparoscopic Radical Prostatectomy.” It was published in European Urology in 2012, and has been cited 21 times, according Thomson Scientific’s Web of Knowledge. Here’s the correction note (paywalled — tsk, tsk):
In the introduction, the Nationwide Inpatient Sample found that 61.3% (not 60%) of radical prostatectomies from October 2008 to December 2009 were robotic assisted. In the third paragraph of the discussion section, the sample size should be 1417 (not 1546), as detailed in Table 1. In Table 1, the positive surgical margin rate should be 7.83% (not 7.9%). Our conclusions are not affected and remain unchanged.
The correction note does not address previous scrutiny of the paper, in the form of a 2012 editorial that was pretty critical of the paper’s conclusion:
To move this forward and in the absence of anatomic evidence that the NVB provides any relevant neural supply to facilitate male urinary continence, the urologic equivalent of the Large Hadron Collider would need to be deployed to demonstrate that the autonomic nerve supply to the penis plays any significant role in the control of male urinary continence.
Tewari and his co-authors responded to the editorial…and then had to retract the response. They had included “mislabeled, non-referenced” images in the rebuttal from a 2005 paper. In 2014, one of the editorial authors, Declan Murphy, told us what it was like to compare the images (emphasis ours):
When we got the original 2005 paper out our jaws hit the floor!! We couldn’t believe they had basically added new labels onto the 2005 paper and had not referenced the paper in the text (making it less easy to spot), and this action totally changed the meaning of the images. By swapping the autonomic nervous system for the somatic, it’s like swapping black for white. It nicely supported the rebuttal of the argument we made in our editorial, but it was based on misrepresentation of other (good) work.
Moving on…the other papers that contain new errata are all published in BJU International. All are included in one erratum note:
This erratum is to clarify baseline missing data, rounding, and a few typographical errors in some of our published tables and figures. The missing data for these noncritical variables were not specifically discussed and typographical errors were overlooked in the editing and proofreading process.
Here’s the text specific to “Assessment of required nodal yield in a high risk cohort undergoing extended pelvic lymphadenectomy in robotic-assisted radical prostatectomy and its impact on functional outcomes,” which has been cited 14 times:
In the Sagalovich et al. [1] paper, Table 1 has some data mismatching in the low- and intermediate-risk patients, which were not the subject of this manuscript, and some typographical errors in the high-risk cases (number mismatch in two patients in Tables 2 and 3).
The note also includes clarifications on “Visual cues as a surrogate for tactile feedback during robotic-assisted laparoscopic prostatectomy: posterolateral margin rates in 1340 consecutive patients,” which has been cited 26 times:
In the Tewari et al. [2] paper, Table 1 should have a footnote stating that 4% of the first 500 patients had missing biopsy data; there was also a typographical error in the first data column, such that the percentage of positive posterolateral surgical margins [PLSM+% (T2–T4)] should read 2.2% rather than 2.1%. In Table 5, the last data column should read 83.6% for the T2 rate (instead of 83%) and 9.4% for the positive surgical margin (SM+) rate (instead of 10%).
And, “Total reconstruction of the vesico-urethral junction,” which has been cited 101 times:
In the Tewari et al. [3] paper, Table 1 should have a footnote stating that 11% of anterior reconstruction patients had missing biopsy data, 4% for the pathological Gleason data, and 1% of the total reconstruction patients had biopsy data missing.
And, “Anatomical grades of nerve sparing: a risk-stratified approach to neural-hammock sparing during robot-assisted radical prostatectomy (RARP),” cited 32 times:
In the Tewari et al. [4] paper, the first box in Figure 1 should read that 2317 patients were >12 months from the surgery date and were eligible for inclusion. Table 1 should have a footnote stating that only 1198 patients had complete information (as listed in the results).
These data are relevant to another paper, “Nerve sparing can preserve orgasmic function in most men after robotic-assisted laparoscopic radical prostatectomy,” cited eight times:
These changes will help readers in understanding the tables and missing data handling even for the noncritical variables [5]. Our conclusions and message remains unchanged.
The note includes a statement from the BJUI Statistical Editor:
Tewari and colleagues have submitted errata to a number of their papers after an eagle-eyed reader identified potential errors in the numbers reported in the tables. In the first paper, for example, the errors arose due mainly to inconsistent handling of missing data but the main message of the paper remains valid. Nonetheless, this does highlight the need for accuracy in reporting. The only proxy readers have for the rigour of any study is the published report, and, whilst occasional errors are inevitable it is the responsibility of both authors and the editorial process to minimise these so that readers can have confidence in findings.
Terwari told us why the changes — which he called “clarifications” — were made:
As mentioned in the editorial comment, one reader needed explanations for some of the missing data in some of non critical baseline variables in the tables and typos. While the exact numbers were mentioned in the tables, further clarification was provided after we reviewed the manuscripts with the editors and went through a re-review process by a bio statistics editor of the journal (who wrote the editorial). It is important to note that none of these clarifications impacted the outcomes measured or the findings of our study.
At the moment, the note is linked to just the first article. We asked the journal why, and managing editor Scott Millar told us:
Thank you for bringing it to our attention that the erratum has only been applied to the first of these five articles. This is an error and I will be working with the publisher to rectify this as soon as possible.
We won’t be making any further statements regarding these articles or the erratum.
Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, and sign up on our homepage for an email every time there’s a new post. Click here to review our Comments Policy.
I wonder if the European Urology editors have had any contact with the BJUI editors? In cross-journal and cross-publisher cases like these, surely it makes sense for editors to come together to discuss details?