Abstract
Background: The author examined the association between better
clinical monitoring and the development of aminoglycoside toxicity
among patients who received aminoglycosides as treatment for
drug-resistant tuberculosis in a retrospective cohort study.
Original data were obtained via chart abstraction for patients
admitted to the tuberculosis unit of the University of Texas Health
Science Center in Tyler, TX between 1985 and 2010. The toxicities
of interest were ototoxicity (measured by serial audiometry) and
nephrotoxicity (measured by serial serum creatinine
determinations).
Methods: Multivariate logistic regression was used to examine
the effect of increased frequency of clinical monitoring on
decreasing toxicity. Count of total audiograms and serum
chemistries and total regimen changes were used as proxies for
clinical monitoring. Stratified analysis, likelihood ratio test,
and backward elimination methods were used to examine effect
modification by risk factors previously established in literature,
using corresponding interaction terms, while change-in-estimate was
used to assess confounding.
Results: In total, 541 observations within the original
dataset were treated with aminoglycosides; 420 had information on
nephrotoxicity development, and 93 had information on ototoxicity
development. Forty-four percent of the ototoxicity dataset (n=93)
developed ototoxicity (P=0.254) while 23 % of the nephrotoxicity
dataset (n=420) developed nephrotoxicity (P<0.0001).
Multivariate logistic regression showed significant associations
between both aminoglycoside toxicities; ototoxicity and audiogram
frequency (crude OR 1.35, 95% CI 1.16,1.58), and nephrotoxicity
with creatinine measurement frequency and regimen change frequency
(crude OR 1.41, 95% CI 1.29,1.54).
Conclusion: While odds ratios were not in the expected
direction, they do not completely contradict the original
hypothesis due to use of the proxy variables and lack of a time
component. Attending physicians might have noticed developing
toxicities and ordered more toxicity measurements and dose changes
to manage them rather than ordering more of these prior to the
development of the toxicities
Table of Contents
INTRODUCTION 1
MATERIALS AND METHODS 4
Data 4
Statistical Analysis 4
RESULTS 7
Univariate descriptive statistics 7
Bivariate analysis 8
Multivariate logistic regression analysis 9
DISCUSSION 11
Strengths 13
Limitations 13
Future directions and public health importance 14
Conclusion 14
References 16
Tables 19
Table 1. Socio Demographic Characteristics of 541 Patients Treated
with Aminoglycosides for Persistent Tuberculosis in Tyler, TX; 1985
- 2008 19
Table 2. Aminoglycoside Characteristics of 541 Patients Treated
with Aminoglycosides for Persistent Tuberculosis in Tyler, Texas;
1985 through 2008 21
Table 3. Audiometry Related Characteristics of 541 Patients Treated
with Aminoglycosides for Persistent Tuberculosis in Tyler, TX; 1985
- 2008 23
Table 4. Serum Creatinine Related Characteristics of 541 Patients
Treated with Aminoglycosides for Persistent Tuberculosis in Tyler,
TX; 1985 - 2008 24
Table 5. Crude Associations of Study among 541 Patients Treated
with Aminoglycosides for Persistent Tuberculosis in Tyler, TX; 1985
- 2008 25
About this Master's Thesis
Rights statement
- Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
School |
|
Department |
|
Degree |
|
Submission |
|
Language |
|
Research Field |
|
关键词 |
|
Committee Chair / Thesis Advisor |
|
Committee Members |
|
Partnering Agencies |
|