How to Read: A new equation to estimate glomerular filtration rate. [Reading Guide]

This guide will help you efficiently read "A new equation to estimate glomerular filtration rate" by Levey et al., published in Annals of Internal Medicine in 2009, using the three-pass approach.

First pass (5–10 min)

What to skim in this paper: Begin by quickly scanning the following sections to get a high-level overview:

• Title: "A New Equation to Estimate Glomerular Filtration Rate" – This immediately tells you the paper's core contribution.
• Abstract (Page 2): Read the entire abstract. Pay attention to the Background, Objective, Design, Results, Limitations, and Conclusions. This section provides a concise summary of the entire study.
• Introduction (Page 2, last paragraph): Focus on the problem statement and the paper's objectives. Note the limitations of the existing MDRD Study equation.
• Discussion (Page 7, first and last paragraphs): Read the opening statement summarizing the findings and the final concluding sentence.

What to notice:

• Contributions: The paper introduces the CKD-EPI creatinine equation, a new method for estimating Glomerular Filtration Rate (GFR).
• Problem: Existing GFR estimation equations, particularly the MDRD Study equation, have limitations, including imprecision and systematic underestimation of GFR at higher levels.
• Setting: The study aims to improve GFR estimation for routine clinical use and public health assessments of chronic kidney disease (CKD) prevalence in adults.
• Key finding (from Abstract): The CKD-EPI equation is more accurate than the MDRD Study equation, especially at higher GFRs, and could replace it.

Second pass (up to 1 hour)

Using SpatialRead in the Second Pass

• Use the Screenshot tool to capture figures, tables, diagrams, and formulae from the PDF and use AI for simple explanations.
• Ask: "Explain Figure X at a high level", "Compare Table 2 vs 3", or "What does the y-axis represent here?"
• Download SpatialRead →

What sections/figures to focus on: In this pass, you'll delve deeper into the methodology and key results to understand the study's claims and evidence.

• Abstract (Page 2): Re-read for a more detailed understanding of the study design, patient numbers, and main quantitative results.
• Introduction (Page 2): Read the full introduction to grasp the clinical significance of GFR estimation and the specific shortcomings of the MDRD equation that the CKD-EPI aims to address.
• Methods (Pages 3-4):
  • Equation Development and Validation: Understand the overall strategy, including the use of separate development and validation datasets.
  • Data sources: Note the number of studies and participants (e.g., 10 studies for development, 16 for external validation, 16,032 for prevalence estimates).
  • Laboratory methods: Understand how serum creatinine was recalibrated.
  • Metrics for equation performance: Pay close attention to how bias, precision, and accuracy were assessed (median difference, interquartile range, P30, root mean square error).
• Results (Pages 5-6):
  • Table 1: Subject Characteristics by Dataset (Page 13): Skim this table to understand the demographic and clinical characteristics of the study populations (development, internal validation, external validation). Notice any significant differences between datasets.
  • Description of CKD-EPI Creatinine Equation (Page 5): Read this section to understand the variables included in the new equation and how it differs from the MDRD equation (e.g., spline for log serum creatinine).
  • Table 2: The CKD-EPI Equation for Estimating GFR on the Natural Scale (Page 14): This is a critical table. Understand the structure of the equation and how it varies by race, sex, and serum creatinine levels.
  • Figure 1: Comparison of performance... (Page 12): Analyze the two panels. Panel 1 shows measured vs. estimated GFR, and Panel 2 shows the difference. Observe visually how the CKD-EPI (left panel) reduces bias compared to MDRD (right panel), especially at higher GFRs.
  • Table 3: Comparison of Performance... in the Validation Dataset (Page 15): This table quantifies the improvements. Focus on the "All Patients" column and the "Patients with estimated GFR ≥60 ml/min/1.73 m²" column. Note the improvements in median difference (bias), interquartile range (precision), and P30 (accuracy) for CKD-EPI.
  • Table 4: Comparison of the CKD-EPI and MDRD Study Equations in Estimating GFR Stage... (Page 16): Examine how the new equation reclassifies GFR stages, particularly the shift to higher values.
• Discussion (Pages 7-8): Read the full discussion to understand the authors' interpretation of the results, the implications for clinical practice and public health, and the study's strengths and weaknesses.

Key terms/questions to note:

• What specific variables are used in the CKD-EPI equation?
• How does the CKD-EPI equation handle serum creatinine differently from MDRD (spline term)?
• What are the quantitative improvements in bias, precision, and accuracy of CKD-EPI over MDRD?
• How does the new equation impact the estimated prevalence of CKD in the U.S. population?
• What are the main advantages and disadvantages of the CKD-EPI equation as identified by the authors?

Third pass (1–5 hours)

Using SpatialRead in the Third Pass

• Highlight complex keywords, dense paragraphs, equations, or proofs and ask for a plain-English explanation that preserves notation.
• Use AI to explain the concepts in a way that is easy to understand.
• Ask for alternative phrasings or analogies to confirm understanding without changing technical meaning.
• Connect concepts with each other using AI to build a comprehensive understanding of the paper.
• Download SpatialRead →

Deep reading guidance: This pass involves a critical and thorough examination of every detail, including supplementary information (if available in the full publication) and references.

• Methods (Pages 3-4):
  • Scrutinize the "Analyses in the development dataset" and "Analyses in the external validation dataset" subsections. Understand the statistical methods used, including linear regression, smoothing splines, and criteria for including variables.
  • Pay attention to the specific statistical tests mentioned (e.g., bootstrap methods for confidence intervals, McNemar test for concordance).
  • Understand the process of "Estimation of U.S. prevalence" and how NHANES data was used.
• Results (Pages 5-6):
  • Go through all tables (Table 1, 2, 3, 4) and figures (Figure 1, 2A, 2B - though 2B is only referenced in text in this PDF) in detail. Understand the meaning of all reported statistics and p-values.
  • For Table 2, ensure you can apply the equation given different patient characteristics.
  • For Table 3, analyze the confidence intervals for each metric to understand the certainty of the reported improvements.
  • For Table 4, trace how specific patient groups are reclassified between GFR stages by the two equations.
• Discussion (Pages 7-8):
  • Critically evaluate the authors' interpretations. Do the results fully support their conclusions?
  • Focus on the "Limitations" section (Page 8). Understand the weaknesses of the study, such as limited representation of certain populations (elderly, specific minorities) and the inherent limitations of serum creatinine as a marker.
  • Consider the implications for future research, such as the need for novel filtration markers.
• References (Pages 9-11): Skim the references to identify foundational papers (e.g., MDRD Study equation, GFR assessment guidelines) and related work.

Replication pointers if applicable: While full replication of such a large-scale study is complex, you can assess the reproducibility of key aspects:

• Equation Derivation: The paper details the statistical modeling approach (linear regression, splines) and the variables used. If you had access to similar raw data and the specified statistical software, you could attempt to reproduce the coefficients for the CKD-EPI equation (Table 2).
• Performance Metrics: The methods for calculating bias, precision, and accuracy (median difference, IQR, P30, RMSE) are clearly defined. With a validation dataset and the equations, you could calculate these metrics.
• Data Sources: The paper lists the types of studies and populations included (e.g., Medline database, CKD-EPI collaborators, NHANES). The Appendix (not included in this PDF) would likely contain more specific details on these sources.
• Laboratory Methods: The recalibration of serum creatinine to the Roche enzymatic method is described, which is crucial for standardizing inputs.