Methods
We queried the National Inpatient Sample (NIS) (2007 to 2014) a
nationally weighted data of all hospital discharges in the United States
to describe the frequency of various complications in AIS patients with
TTS and also to determine the predictors of all-cause in-hospital
mortality. The NIS is a part of the Healthcare Cost and Utilization
Project (HCUP) funded by the Agency for Healthcare Research and Quality
(AHRQ) [11]. This data consists of a sample from 20% of total
hospitalizations from nonfederal US community hospitals representing
95% of the general population in the US. The NIS contains de-identified
data; hence, our study was exempt from an institutional review board
authorization.
We used International Classification of Diseases, 9th Revision, Clinical
Modification (ICD-9-CM) to identify primary admissions for AIS (433.01,
433.11, 433.21, 433.31, 433.81, 433.91, 434.01, 434.11, 434.91 and 436)
and then using ICD-9 CM code 429.83 admissions with TTS as a secondary
discharge diagnoses were filtered.
The primary outcome of our study was to assess the prevalence, trends
and all-cause inpatient mortality in AIS cohort with TTS vs. without
TTS. Secondary outcomes included frequency of in-hospital complications
such as cardiogenic shock, venous thromboembolism, respiratory failure,
need for mechanical ventilation and healthcare resource utilization
[discharge disposition, mean length of stay (LOS) and hospital
charges] among AIS-TTS cohort as compared to AIS- non-TTS cohort.
We utilized designated weight values to produce nationally
representative estimates and survey multivariable logistic regression
models were performed after adjusting patients and hospital-level
confounders to estimate the odds and 95% confidence intervals (CI) of
in-hospital mortality and other complications in AIS cohort with TTS as
compared to without TTS. We used SPSS v22 (IBM Corp., Armonk, NY) to
perform all statistical analyses.