Review History

To increase transparency, PeerJ operates a system of 'optional signed reviews and history'. This takes two forms: (1) peer reviewers are encouraged, but not required, to provide their names (if they do so, then their profile page records the articles they have reviewed), and (2) authors are given the option of reproducing their entire peer review history alongside their published article (in which case the complete peer review process is provided, including revisions, rebuttal letters and editor decision letters).

New to public reviews? Learn more about optional signed reviews and how to write a better rebuttal letter.


  • The initial submission of this article was received on November 25th, 2014 and was peer-reviewed by 2 reviewers and the Academic Editor.
  • The Academic Editor made their initial decision on January 13th, 2015.
  • The first revision was submitted on January 29th, 2015 and was reviewed by the Academic Editor.
  • A further revision was submitted on February 12th, 2015 and was reviewed by the Academic Editor.
  • The article was Accepted by the Academic Editor on February 13th, 2015.

Version 0.3 (accepted)

· · Academic Editor


I suggest "systematic" rather than "systemic" here: "No systematic evaluation of tic disorders or chronic somatization was performed ..."

Version 0.2

· · Academic Editor

Minor Revisions

Dear Drs. Bastian,

This "minor revisions" notice will clear the system up so you can upload the amended files with the revisions we discussed.

Kevin Black MD

Version 0.1 (original submission)

· · Academic Editor

Major Revisions

This is an interesting manuscript with potential clinical utility. However, one of the reviewers identifies significant concerns with the strength of the data. After reflection, I believe that the manuscript may add to the literature if the authors can adequately tone down the conclusions and more fully acknowledge the study's limitations.

The data are what they are, and may be unique, but the manuscript touches at most lightly on the potential impact of the concerns noted by Reviewer 2. The authors also assert that the medications studied caused the relief of symptoms observed and that "each medication has its place" in treatment. As the authors note, the long duration of prior symptoms lends some support to the idea that without intervention symptoms would have persisted. Nevertheless, the mechanism of improvement cannot be known, especially whether the improvement was a placebo effect. In other words, we don't know whether they would have improved to a similar extent with other interventions (excluding antihistamines, asthma inhalers, and PPIs). At lines 132-133, the authors assert that previous studies have used a randomized, placebo controlled design, but in fact the cited study that used tricyclics was not controlled, and the controlled study used another drug class. The concern of a placebo effect is especially relevant given the extremely low doses used for amitriptyline (40mg) and desipramine (25mg) (median final doses in responders from Table 4). These doses certainly don't work by whatever mechanism makes them effective for depression.

I would add a request that the authors address a couple of concerns about differential diagnosis. Cough is a very common simple phonic tic. Most tic patients describe premonitory sensations (like a tickle in the throat before a nonproductive cough), and many patients report worsening of symptoms with various infections. Some tic patients also report onset of a new tic after repeated non-tic movements or vocalizations, for instance onset of a coughing tic after frequent non-tic coughs during a viral URI. As far as we can tell from the data provided (first 2 sentences of Results), some patients reported here may have had onset of symptoms before age 21, in which case all the clinical features used here to define sensory neuropathic cough would be absolutely typical of a tic disorder. Doubtless tic history could not have been adequately investigated prospectively in each of these patients in a busy ENT practice, but this should be acknowledged as another possible consideration in differential diagnosis.

Similarly, a patient who presented with undiagnosed persistent cough _and_ a history of many other unexplained medical symptoms since youth might more parsimoniously be diagnosed with somatization disorder. Ruling this out generally takes a good bit of medical history and a prospective search, but this should also be addressed in terms of patient selection for future research or clinical practice.

Finally, the possibility that these symptoms represent a sensory neuropathy is very interesting, and the authors summarize data suggesting that possibility. It is not clear however from the discussion whether there is any direct evidence in these patients of damage to sensory nerves (clinical electrophysiology, biopsy, autopsy). This point should be addressed since the assumption of sensory neuropathy pervades the article from the title on.

Reviewer 1 ·

Basic reporting

This paper provides a clear introduction to the causes of chronic cough, shifting on to cover the less commonly encountered SNC; the second introductory paragraph is particularly useful as the clinical syndrome of SNC is expanded upon. This is beneficial for clinicians who are not familiar with such alternative causes of cough or have come across SNC under different terminology.
The authors have used clear English.

Experimental design

Authors make it very clear this is a retrospective study design. Consideration has been given to the diagnostic credibility of those admitted to the study, with sole inclusion of those who have been unsuccessfully treated for alternative causes of cough. The other inclusion criteria were also appropriate.
Ethics approval was obtained from the institutional review board of Advocate Healthcare.
Method of recording symptom reduction has not been validated but appropiate given retrospective nature of the study.

Validity of the findings

Relevant demographics included. Clear, valuable tables tabulating symptom reduction as well as treatment trials. Appropriate conclusions are drawn from the orginal findings.

Comments for the author

This is an interesting and well written original study.


Basic reporting

No Comments

Experimental design

There are serious methodological errors in the treatment protocol that render it invalid for publication.

In Table 1, Step 1: First-line medication trial, Trial method section, the instructions read: “Take 10 mg daily, two hours before bed, for the first day or two. Then, until symptoms are reduced 80% or more, or until a maximum of 80 mg a day is reached, continue increasing the daily dose by 10 mg every one or two days.”

In Table 1, Step 2: Phone follow-up, the text reads, “either 14 days after starting the medication or sooner if desired.”

In Table 1, Step 2: Phone follow up, the text reads, “Physician tells the staff to instruct the patient to continue increasing the dose (if there is a partial response), to adjust the time of day or other logistics of use, or to taper off current medication and then begin another medication (if potential of current medication seems to have been exhausted).

Simply put, the stated treatment protocol is not standardized and is not able to be replicated by others, either in a replicated study or in clinical practice. There are four uses of the word “or” in Step 1 of the treatment protocol. It is impossible to replicate this. In Step 2, it is written “to adjust the time of day or other logistics of use”. What are the logistics of use that can be adjusted. Again, this protocol is not able to be replicated. Also, it is written, “if potential of current medication seems to have been exhausted”. What criteria are used to determine this and how can any other practice or researcher replicate this protocol if these criteria are not explicitly stated?

Validity of the findings

The primary outcome measure of this analysis was self-reported percentage of cough reduction. This is not a valid measure of clinical response to treatment, as the authors state. It is not possible to compare one patient’s perception of percentage reduction in cough to another’s and arrive at a conclusion regarding efficacy of any treatment, especially when the raw data of the patients presented for analysis is analyzed. The cohort included patients with 1-200 episodes of cough per day. Any reduction in cough would be a 100% response for the patient with one episode per day; however, a reduction of one episode a day for a patient with 200 episodes a day would be a response of only 0.05%. A true analysis could have been performed by tracking cough counts, before, during and after treatment, and analyze actual percentage reduction in each patient. This could have been done, especially if number of episodes per day, frequency and severity prior to treatment were all collected at the initial visit.

There is no discussion in the manuscript regarding the lack of statistical analysis of the treatment methodology. I did not see any discussion of a power analysis, so it is not clear that the sample size was sufficient to demonstrate an effect of any of these treatments. There is no discussion regarding the comparative efficacy of the medications used, other than the percentage of patients demonstrating response to each medication. Furthermore, the treatment medications are compared to each other using the “average percentage reduction” at final dose of medication. Using the average of a dataset eliminates the possibility of demonstrating any effectiveness, as average, statistically, is the mean of a data sample.

There was no discussion regarding the lack of comparative statistical analysis between the treatment medications. It is not possible to state that the medications have a similar rate of efficacy without providing some statistical proof that the data are similar. Critically, this cannot be stated when the numbers of patients exposed to each treatment medication are not equal. As shown in Table 4, eighteen patients were treated with amitriptyline, fifteen with desipramine, and twelve with gabapentin. Patients treated with gabapentin had the largest rate of response at 83%, but is this statistically the same or different than the 73% of patients who had a response to desipramine?

Comments for the author

I read, with great interest, the above manuscript, regarding the treatment of sensory neuropathic cough. This is an interesting clinical entity, with some significant phenotypic overlap with chronic vocal tic. I was most interested in reviewing this manuscript as the age of the patients described was significantly older than those who would be diagnosed with vocal tic, and the role of specific environmental triggers inducing this entity is not seen in vocal tic. Additionally, I applaud the authors for attempting to establish a treatment protocol for this troubling condition.

The inclusion/exclusion criteria for the cohort were very stringent and clear, leading to a homogenous population with no confounding diagnoses.

Despite these positive features, there are a number of methodological problems with this manuscript that preclude it from being acceptable for publication.

All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.