What grace may be added to commonplace matters by the power of order and connection.
A review article is a compilation of previously published thoughts and observations by multiple authors. In structure, it shares features with encyclopedias but is narrower in scope, shorter in length, more timely, and usually directed toward the specific audience of the journal in which it is published. The above quotation by Horace, which appeared on the title page of the Encyclopdie edited by d'Alembert and Diderot in the 18th century, sums up the value and power of this type of work.
Value of a Review
In the construction of a compilation, the order of presentation of the material can enhance the understanding of any connections. For example, many encyclopedias use an alphabetic sequence as the structure, whereas journal review articles such as this one are arranged topically, usually in order of a condition's frequency or sometimes with what is best understood discussed first. To ensure maximum clarity, topics and all critical terms should be defined and used consistently throughout the manuscript. While easy in concept, this is difficult to put in practice given the various definitions or lack thereof in the many manuscripts from which the article is derived.
After reviewing and understanding the literature and after testing the published observations against the author's own experience, the task of cataloging and including (or not including) articles begins. From the reader's perspective, it is important to understand the concepts underlying the catalog, what was included, and what was not included. Inherent to the cataloging process is the concept of connections between topics and entities and the converse—the lack of connections between other topics and entities.
Most articles in clinical dermatology, especially concerning rare events are observational in nature and confounded by the unique situation of a particular patient or group of patients, and sometimes by the practices of a specific clinic or institution. It is only by collating similar observations from multiple sources that we can approach an understanding of the underlying biologic principles.
Structure of the Article
The review article by Agha et al concerns dermatologic problems in cancer patients and, as such, crosses the boundaries of dermatology, oncology, and primary care medicine. The scope is predominantly the adverse effects of therapy for internal organ cancer, with lesser emphasis on the cutaneous manifestations of the cancer itself. The article is organized according to the morphologic appearance of the skin lesions, which in two flowcharts is nicely connected with the most frequent causal agents. For easy searchability, these flowcharts are superb. Therapy of each cutaneous adverse effect is provided in general terms, but for specific directions the supplied references must be read.
Where Does This Review Article Lead?
For the clinician with a specific patient, it is first a matter of recognizing the dermatologic problem and the probable association with a specific causal agent. Since the underpinning of medical diagnosis is based on probabilities and context, having more information from what has been published in the literature would be helpful. For example, knowing that 6% to 42% of patients treated with cytarabine, fluorouracil, doxorubicin, or methotrexate develop hand-foot syndrome is helpful. However, providing the numerators (affected) and denominators (exposed) in the published reports for each drug would offer more powerful information. Dennis Frisman has successfully used this tactic of compiling information in developing Immunoquery, which provides an analysis of positive and negative immunostains in tumors as well as the percentages of specific tumors that are positive with a particular immunostain or combinations of immunostains.
Following recognition of the specific adverse effect, the clinician must solve the problem of management. Again, more specific and critical information in the article would be helpful. While few controlled clinical trials have investigated these therapies, the authors could have provided their expertise on the rank order of therapies, how they should be best used, and the expected endpoints. The overall goals of therapy are to enhance quality of life by managing the adverse effects as well as to achieve optimal outcomes from a given therapeutic agent.
For the clinician-scientist, understanding the mechanisms underlying the development of adverse effects is important. Such understanding hopefully will lead to rational intervention for prevention or amelioration. Knowing the interconnections between the agents and adverse effects may lead to clues. For example, mucositis can be caused by radiation, dactinomycin (Cosmegen), or bleomycin; perhaps interconnections can be drawn with bleomycin also causing neutrophilic hidradenitis and sclerosis. Indeed, bleomycin can adversely affect the oral mucosa with erosions, sweat glands with a neutrophilic inflammatory response, and dermal connective tissue with collagen production.
Is there a common pathway, multiple pathways, or other required conditions for specific adverse events to occur? Moreover, why doesn't the adverse event occur in all patients given the same diagnosis, dose, and mode of delivery? It is just as important to ask why something does not occur as well as why it does. Compilations of existing information can lead to these types of questions and directions for research.
In conclusion, there is significant value in the compilation of information, but a literature review should go far beyond simply a recording of existing information, as it provides a means of solving specific clinical problems as well as a tool to expand knowledge. Knowledge must be recorded, categorized, analyzed, and then made easily available to be useful.
Financial Disclosure: The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.
1. STATdxPathIQ/ImmunoQuery home page: Available at https://my.statdxpathiq.com/PathIQ/Login.do. Accessed October 12, 2007.