This book intends to equip you with a foundational knowledge of landmark articles in general surgery. The introductory chapter, “Finding Answers,” provides tools with which to conduct literature searches and identify high-impact articles. In this chapter, we describe a method for reading and interpreting journal articles in order to apply them to your clinical practice. Whether you are reading a landmark study within a field, running a literature search to answer a clinical question, or scanning an interesting abstract in the latest issue of a journal, assessment of the literature requires a logical and systematic approach to reading scientific manuscripts. The following discussion outlines key questions to consider when reading each section of a journal article.

1. 
   

   Title and abstract

   
   

   Should I read this paper?

   
   

   Despite your new-found skills in literature searching, PubMed often produces lists of articles that vary in relevance to your clinical question. A good title and abstract should hint whether the study is worth your time to read.

   
   

   Abstracts are insufficient, however, to answer any clinical question, and should be used only to gauge whether to spend time reading the full article. Remember that the abstract is a 200–300-word advertisement for the paper. It is designed as much to hook the reader as to summarize the study. The abstract’s conclusion section will provide one or two sentences that usually showcase the most impressive results, and most abstracts will avoid stating negative results.

   
   

   For example, a study evaluating a new imaging test may reveal that the test has very poor sensitivity and specificity in the general population, but works well in a narrowly defined subset. The abstract’s conclusion may state only the positive findings, ignoring the narrow range of applicability. The abstract will not reliably tell you whether the article’s conclusions are generalizable, applicable to your patient population and practice, or conclusive. To answer such questions, you need to continue reading.

   
   
2. 
   

   Introduction

   
   

   Why was this study done?

   
   

   The introduction section orients you to the study’s topic with relevant citations to other important papers in the field. If you are not familiar with the field, the introduction is particularly helpful, and it serves as a good starting point in a search for landmark articles. Finally, the introduction provides insight into the authors’ motivations for writing the article.

   
   
3. 
   

   Methods and results

   
   

   The methods and results are the most important sections of the paper. These are the only sections that, taken alone, can answer a clinical question. If you want to read only two sections, after deciding a paper is relevant on the basis of its title and abstract, those sections should be the methods and results.

   
   

   Does this paper apply to my patient?

   
   

   Clinical studies generally present a description of the included patient population in “Table 1.” This table provides the overall picture of the included population, and if the study has more than one group, it will indicate whether these groups have similar baseline characteristics (gender, age, comorbidities).

   
   

   In the text, the authors describe the inclusion and exclusion criteria used to choose the included population. These criteria can reveal the relevance of a study to your patient. For example, the study by Blakely et al¹ evaluating interval appendectomy for perforated appendicitis included only patients less than 18 years of age. If you are treating a 65-year-old man with multiple comorbidities who presents with perforated appendicitis, the Blakely study may not apply to your patient.

   
   

   Are these results generalizable to my practice?

   
   

   In addition to evaluating the study population, it is also important to evaluate the study method for generalizability. A study evaluating a new treatment does so within the context of a larger clinical setting. If the larger setting does not reflect your practice, the results may not be generalizable. For example, Popadich et al² showed that patients have lower reoperation rates with no increase in complications with routine central lymph node dissection for cN0 papillary thyroid cancer. The included hospitals, however, were high-volume centers with specialized surgeons. If you are working at a community hospital without endocrine specialists, the results may not apply to your practice.

   
   

   What is the study design?

   
   

   The methods section should also allow you to describe the study’s design. Understanding the study design is an important step toward understanding the study’s level of evidence, potential limitations, and statistical outcomes. In general, studies fall into two major categories: descriptive and analytic. Descriptive studies do not try to quantify a relationship, but rather to describe the prevalence, incidence, or experience of an entire group. These studies provide a description of the problem, but do not attempt to provide a cause-and-effect analysis. Analytic studies attempt to quantify the relationship between two factors, usually between an exposure or intervention (for example, early goal-directed therapy) and an outcome (mortality). Analytic studies can be either observational or experimental. The major types of study designs within these categories are summarized in Table 1.^3,4

   
   
4. 
   

   Discussion and conclusion

   
   

   What are the limitations to this study?

   
   

   The methods and results should allow you to recognize the study’s main limitations, especially as you identify the study design. The discussion should acknowledge these limitations explicitly, with the authors’ comments. Every study faces a tension between generalizability and reliability. To be generalizable, a study needs to represent the real world; it needs to include patients with comorbidities and poor compliance in a variety of practice patterns. With all of those unknown variables, however, the study is at risk for confounders (see Box 1). To reduce confounders, authors try to make the population and study setting as restricted as possible. This makes it less likely that unknown confounders are impacting the outcome, but it also makes the study results less generalizable. These limitations are understandable and unavoidable, but they are important to identify when deciding if a study applies to your patients.

   
   

   Where do the authors think the field is headed?

   
   

   The discussion section will also help you understand the authors’ views of the paper’s relevance. They will usually explain how the results of the paper can change the field, and identify open questions that can drive further research. It is worth noting that the discussion section gives you only the author’s opinions. Often, for a large clinical trial or particularly interesting paper, other researchers and clinicians will share their opinions as well. Many of the articles in this book sparked letters to the editor or even editorials within the same journal issue. When you access an article on a journal’s website, “related articles” will almost always appear on the top or side margin. These related articles often include letters to the editor, editorials, or commentaries that can give you some understanding of how the study was received by other experts in the field. For example, the Rivers Trial on early goal-directed therapy for septic shock was published in the November 8, 2001 issue of the New England Journal of Medicine (NEJM). In that same issue, an editorial was published on hemodynamic and metabolic therapy in critically ill patients that provides significant context and critique of the Rivers Trial.⁵

   
   

   Will this change my practice?

   
   

   The final question to ask of any clinical study is whether the results of this study change or confirm your practice. All of the questions you ask yourself throughout the article culminate in this decision. As a result of this trial, will you prescribe (or stop prescribing) a drug? Will you change the way you perform a procedure? Will you continue to require a certain imaging test, or stop requiring an unnecessary test? This is the clinical judgment call every practicing surgeon makes: When reading a new study, how will you integrate the information into your existing knowledge base, and what choices will you make for your patients? This is our obligation as clinicians, and the fundamental reason why literature searches, journal articles, and evidence-based practice is so important.